[BP-145] htt:tTask in PRD-03 with htt:tTaskDetails

Created: 09/Jun/10  Updated: 24/Jun/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Mark Ford

During the course of work, Mark Ford identified the following issue

... do you know what the following API elements
should be referring to instead of htt:tTask? I'm guessing that htt:tTask
was from an older version of the schema and was renamed to something else.
These elements are from ws-humantask-api.wsdl.

     <xsd:element name="getParentTaskResponse">
       <xsd:complexType>
         <xsd:sequence>
           <xsd:element name="parentTask" type="htt:tTask" minOccurs="0"/>
         </xsd:sequence>
       </xsd:complexType>
     </xsd:element>

     <xsd:element name="getSubtasksResponse">
       <xsd:complexType>
         <xsd:sequence>
           <xsd:element name="subtask" type="htt:tTask" minOccurs="0"
maxOccurs="unbounded"/>
         </xsd:sequence>
       </xsd:complexType>
     </xsd:element>

Dieter reviewed this and provided the following in return.

However, Mark Ford discovered an error in the API WSDL. The type of the
return value of the two operations is wrong because it is not defined in
the data types XML Schema (prefix htt). As a consistency issue between WSDL
and XSD, this was not recognized by the XML Schema validations. Just from
the WSDL point of view, it could be handled as an editorial fix and done
immediately (change "htt:tTask" to "htt:tTaskDetails" in these two places),
but there is also a remark in the API definition table in 7.1.1 that needs
to be changed (for the getSubtasks operation, change "Returns all sub tasks
of a task (created instances + not yet created sub task definitions)") to
"Returns all sub tasks of a task (created instances)".

I propose that we accept Dieters issue.

Proposal:

(1) For the getParentTask and getSubtasks API operations, change the type of the return value from "htt:tTask" to "htt:tTaskDetails"
(2) Change the remark in the API definition table in 7.1.1 that needs to be changed - for the getSubtasks operation, change "Returns all sub tasks of a task (created instances + not yet created sub task definitions)") to "Returns all sub tasks of a task (created instances)".




[BP-144] encoding QNames + Org Entity for HT Advanced query operations

Created: 31/Mar/10  Updated: 14/Apr/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Bug

Priority:

Major

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Mark Ford

Issues raised by: http://lists.oasis-open.org/archives/bpel4people-comment/201003/msg00000.html

Proposal by Dieter Koenig

<dk>
(values in where-clauses - we need to add a clarification here)
(1) For QName instances, we should expect values of the form

      name="{http://example.com}ApproveClaim"

(2) For tOrganizationalEntity instaces, we must split the complex value into separate simple-typed values and should expect them in the form

      potentialOwner.user="userid"
      potentialOwner.group="groupid"
      potentialOwner.user IN ( "userid1", "userid2" , ..., "useridN" )
      potentialOwner.group IN ( "groupid1", "groupid2", ..., "groupidN" )

which can be combined using AND / OR in the usual way.
</dk>

There was a further clarification raised during the 20100331 call. The proposal was amended as follows:

 Proposal: as described above with the modification that the {namespace} part for QName values is optional and treated as wildcards if not specified.




[BP-143] MessageChoice Name Attribute

Created: 29/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter König

Target:

WS-HumanTask 1.1 CD 06 and PRD 01, section 5.3 Message Schema and Appendix B. WS-HumanTask Language Schema and ws-humantask.xsd

Description:

The WS-HT Lean Task pseudosyntax in section 5.3 and all Lean Task examples use a "messageChoice" element with a "name" attribute which has no definition in the XML Schema.

This attribute contains one of the possible values of a "messageField"
element, therefore its semantics are more similar to an XML Schema simple type enumeration value.

Finally, the type of this attribute should not be "xsd:NCName" as it holds values of arbitrary XML Schema simple types.

Proposal:

In the WS-HT Lean Task pseudosyntax in section 5.3 and all Lean Task examples, rename the attribute "name" to "value", and add the definition

   <xsd:attribute name="value" type="xsd:anySimpleType" />

to the "messageChoice" element in Appendix B. WS-HumanTask Language Schema and ws-humantask.xsd.






[BP-142] Definition of element tComment is not complete

Created: 27/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: ws-humantask-types.xsd, WS-HumanTask specification, version 1.1, PR 1
 
Issue Description:
In the XML file: Definition of element tComment is not complete. Attributes id, lastModifiedTime and lastModifiedby are missing.
In the specification: Attribute ID is of type "xsd:string". Other IDs are of type xsd:anyURI.
These inconsistencies shall be removed.
 
Proposal:
In the XML file, include the following XML snippet
<xsd:complexType name="tComment">
     <xsd:sequence>
       <xsd:element name="id" type="xsd:anyURI" />
       <xsd:element name="addedTime" type="xsd:dateTime" />
       <xsd:element name="addedBy" type="htt:tUser" />
       <xsd:element name="lastModifiedTime" type="xsd:dateTime" />
       <xsd:element name="lastModifiedBy" type="htd:tUser" />
       <xsd:element name="text" type="xsd:string" />
       <xsd:any namespace="##other" processContents="lax"
             minOccurs="0" maxOccurs="unbounded" />
     </xsd:sequence>
</xsd:complexType>
 
instead of
<xsd:element name="comment" type="tComment" />
<xsd:complexType name="tComment">
  <xsd:sequence>
    <xsd:element name="addedTime" type="xsd:dateTime" />
    <xsd:element name="addedBy" type="htt:tUser" />
    <xsd:element name="text" type="xsd:string" />
    <xsd:any namespace="##other" processContents="lax"
             minOccurs="0" maxOccurs="unbounded" />
  </xsd:sequence>
</xsd:complexType>
 
 
In the specification (p. 30), include the following XML snippet:
<xsd:complexType name="tComment">
     <xsd:sequence>
       <xsd:element name="id" type="xsd:anyURI" />
       <xsd:element name="addedTime" type="xsd:dateTime" />
       <xsd:element name="addedBy" type="htt:tUser" />
       <xsd:element name="lastModifiedTime" type="xsd:dateTime" />
       <xsd:element name="lastModifiedBy" type="htd:tUser" />
       <xsd:element name="text" type="xsd:string" />
       <xsd:any namespace="##other" processContents="lax"
             minOccurs="0" maxOccurs="unbounded" />
     </xsd:sequence>
</xsd:complexType>
 
instead of
<xsd:complexType name="tComment">
     <xsd:sequence>
       <xsd:element name="id" type="xsd:string" />
       <xsd:element name="addedTime" type="xsd:dateTime" />
       <xsd:element name="addedBy" type="htt:tUser" />
       <xsd:element name="lastModifiedTime" type="xsd:dateTime" />
       <xsd:element name="lastModifiedBy" type="htd:tUser" />
       <xsd:element name="text" type="xsd:string" />
       <xsd:any namespace="##other" processContents="lax"
             minOccurs="0" maxOccurs="unbounded" />
     </xsd:sequence>
</xsd:complexType>




[BP-141] Type of IDs not consistently defined

Created: 27/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: WS-HumanTask specification, ws-humantask-api.wsdl, version 1.1, PR 1
 
Issue Description:
The following IDs are defined to be of type xsd:string:
tTaskAbstract
- id
- parentTaskId
 
tTaskDetails
- id
- parentTaskId
 
tTaskQueryResultRow
- id
- parentTaskId
 
Other IDs are of type xsd:anyURI. All IDs shall be of the same type.
 
Proposal:
Set all IDs in the specification and XML file to be of type anyURI.
 




[BP-140] Task API - batchComplete/batchFail input parameters

Created: 26/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: WS-HumanTask specification, version 1.1, PR 1
 
Issue Description:
batchComplete and batchFail operations are only batch operations that do not have input parameters, other than task identifier. However, the complete operation allows to pass output data of the task. Similarly, the fail operation allows to pass fault data and fault name. Extend the batch operations with optional input parameters.
 
Proposal:
The following changes need to be made in WSDL. The specification text does not need to be changed.
 
Include the following XML snippet:
<xsd:element name="batchComplete">
   <xsd:complexType>
      <xsd:sequence>
         <xsd:element name="identifier" type="xsd:anyURI" maxOccurs="unbounded"/>
         <xsd:element name="taskData" type="xsd:anyType" minOccurs="0"/>
      </xsd:sequence>
   </xsd:complexType>
</xsd:element>
 
instead of
<xsd:element name="batchComplete">
   <xsd:complexType>
      <xsd:sequence>
         <xsd:element name="identifier" type="xsd:anyURI" maxOccurs="unbounded"/>
       </xsd:sequence>
   </xsd:complexType>
</xsd:element>
 
 
Include the following XML snippet:
<xsd:element name="batchFail">
   <xsd:complexType>
      <xsd:sequence>
<xsd:element name="identifier" type="xsd:anyURI" maxOccurs="unbounded"/>
<xsd:element name="fault" type="htt:tFault" minOccurs="0"/>
      </xsd:sequence>
   </xsd:complexType>
</xsd:element>
 
 
instead of
<xsd:element name="batchFail">
   <xsd:complexType>
      <xsd:sequence>
<xsd:element name="identifier" type="xsd:anyURI" maxOccurs="unbounded"/>
      </xsd:sequence>
   </xsd:complexType>
</xsd:element>




[BP-139] Task API - Complete/fail optional parameters

Created: 26/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: WS-HumanTask specification, version 1.1, PR 1
 
Issue Description:
For the complete operation the specification says (on p. 73): "If no output data is set the operation MUST return hta:illegalArgumentFault."
As the data can also be set using operation setOutput, the output data can be defined as an optional parameter. So this constraint should be removed. The output data is already defined as optional in the WSDL file.
 
For the fail operation the specification says (on p. 74): "If fault name or fault data is not set the operation MUST return hta:illegalArgumentFault."
As the fault data and name can also be set using operation setFault, the fault data and name can be defined as optional parameters. So this constraint should be removed. The fault data and name are already defined as optional in the WSDL file.
 
Proposal: (revised below)
Remove the following statements:
On p. 73 (operation complete): "If no output data is set the operation MUST return hta:illegalArgumentFault."
On p. 74 (operation fail): "If fault name or fault data is not set the operation MUST return hta:illegalArgumentFault."

Proposal revision at http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/201002/msg00007.html. Here is the excerpt

Below please find an updated proposal incorporating also comments raised by Dieter and Phil.
 
On p. 73 (operation complete):
Include
"The fault hta:illegalStateFault MUST be returned if the task interface defines non-empty task output but no output data is provided as the input parameter and the task output data has not been set previously, e.g. using operation setOutput."
 
instead of
 
"If no output data is set the operation MUST return hta:illegalArgumentFault."
 
On p. 74 (operation fail):
 
Include
"The fault hta:illegalStateFault MUST be returned if the task interface defines at least one faults but either fault name or fault data is not provided and it has not been set previously, e.g. using operation setFault."
 
instead of
 
"If fault name or fault data is not set the operation MUST return hta:illegalArgumentFault."









[BP-138] Task API - Extend batch operations

Created: 26/Jan/10  Updated: 17/Feb/10

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: WS-HumanTask specification, version 1.1, PR 1
 
Issue Description:
Batch operations allow to apply the same operation with the same data to multiple task, e.g. set the priority of multiple tasks to the same value. However it is currently not possible to apply the same operation to multiple tasks but with different data, e.g. set the priority of multiple tasks to new values, or complete multiple tasks with different output data, etc.
Extend all batch operations to allow applying the same operation to multiple tasks with different data.

 

 Comments 

 

 

Comment by Luc Clement [ 17/Feb/10 10:23 AM ]

closed with no action to take




[BP-137] Task API - Type of commentID not consistently defined

Created: 25/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: ws-humantask-api.wsdl, version 1.1, PR 1
 
Issue Description:
Type of commentID not consistently defined. The addComment operation has input data and output data. The input data defines task identifier to be of type xsd:anyURI. The output data defines comment identifier to be of type xsd:string. Other operations, e.g. updateComment, define comment identifier to be of type "xsd:anyURI".
 
Proposal:
Change comment identifier to be of type xsd:anyURI.
The xsd snippet should look as follows:
 
<xsd:element name="addCommentResponse">
      <xsd:complexType>
- < <xsd:sequence>
   <xsd:element name="commentID" type="xsd:anyURI" />
</xsd:sequence>
     </xsd:complexType>
</xsd:element>
 
In addition, make the following changes:
- use the same naming conventions (ws-humantask-api.wsdl): Change "commentID" to "commentIdentifier"




[BP-136] Task API - Type of task identifier not consistently defined

Created: 25/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: ws-humantask-api.wsdl, version 1.1, PR 1
 
Issue Description:
Type of task identifier not consistently defined. Operations getRendering and getRenderingTypes use task identifier to obtain rendering and rendering types, respectively. They define task identifier to be of type xsd:anyType. Other operations define the identifier to be of type "xsd:anyURI".
 
Proposal:
Change task identifier to be of type xsd:anyURI.
The xsd snippet should look as follows:
 
<xsd:element name="getRendering">
        <xsd:complexType>
          <xsd:sequence>
            <xsd:element name="identifier" type="xsd:anyURI"/>
            <xsd:element name="renderingType" type="xsd:QName"/>
          </xsd:sequence>
        </xsd:complexType>
</xsd:element>
 
<xsd:element name="getRenderingTypes">
        <xsd:complexType>
          <xsd:sequence>
            <xsd:element name="identifier" type="xsd:anyURI"/>
          </xsd:sequence>
        </xsd:complexType>
  </xsd:element>
 




[BP-135] Query operation inconsistencies

Created: 25/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: WS-HumanTask specification and ws-humantask-api.wsdl, version 1.1, PR 1
 
Issue Description:
A couple of inconsistencies have been found:
1. The specification says that getMyTaskAbstract has the following input parameters:

    * task type ("ALL" | "TASKS" | "NOTIFICATIONS")
    * generic human role
    * work queue
    * status list
    * where clause
    * order-by clause
    * created-on clause
    * maxTasks
    * taskIndexOffset

"order-by clause" and "taskIndexOffset" are not included in the wsdl file.
 
2. Operations getMyTaskAbstract and getMyTaskDetails differ in the list of input parameters. Operation getMyTaskDetails does not allow to specify "order-by clause" and "taskIndexOffset".
 
3. The type of "maxTasks" not consistently defined
 
Proposal:
1. Change element getMyTaskAbstract (in ws-humantask-api.wsdl) as follows:
<xsd:element name="getMyTaskAbstracts">
  <xsd:complexType>
    <xsd:sequence>
  <xsd:element name="taskType" type="xsd:string" />
  <xsd:element name="genericHumanRole" type="xsd:string" minOccurs="0" />
  <xsd:element name="workQueue" type="xsd:string" minOccurs="0" />
  <xsd:element name="status" type="htt:tStatus" minOccurs="0" maxOccurs="unbounded" />
  <xsd:element name="whereClause" type="xsd:string" minOccurs="0" />
  <xsd:element name="orderByClause" type="xsd:string" minOccurs="0" />
  <xsd:element name="createdOnClause" type="xsd:string" minOccurs="0" />
  <xsd:element name="maxTasks" type="xsd:int" minOccurs="0" />
  <xsd:element name="taskIndexOffset" type="xsd:int" minOccurs="0" />
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>
 
2. Extend the list of input data of operation getMyTaskDetails to allow defining "order-by clause" and "taskIndexOffset", in the specification (section 7.1.2) and wsdl file .
The xsd snippet should look as follows:
 
<xsd:element name="getMyTaskDetails">
     <xsd:complexType>
        <yxsd:sequence>
         <xsd:element name="taskType" type="xsd:string" />
<xsd:element name="genericHumanRole" type="xsd:string" minOccurs="0" />
<xsd:element name="workQueue" type="xsd:string" minOccurs="0" />
<xsd:element name="status" type="htt:tStatus" minOccurs="0" maxOccurs="unbounded" />
<xsd:element name="whereClause" type="xsd:string" minOccurs="0" />
  <xsd:element name="orderByClause" type="xsd:string" minOccurs="0" />
<xsd:element name="createdOnClause" type="xsd:string" minOccurs="0" />
<xsd:element name="maxTasks" type="xsd:int" minOccurs="0" />
  <xsd:element name="taskIndexOffset" type="xsd:int" minOccurs="0" />
      </xsd:sequence>
   </xsd:complexType>
</xsd:element>
 
3. Change maxTasks to be of type xsd:int instead of xsd:integer for operation "GetTaskHistory".
 




[BP-134] Task API - missing operations in the wsdl file

Created: 25/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: ws-humantask-api.wsdl, version 1.1, PR 1
 
Issue Description:
The following operations are not included in the ws-humantask-api.wsdl file: updateComment, deleteComment, getSubtasks, getSubtaskIdentifiers, hasSubtasks, getParentTask, getParentTaskIdentifier, isSubtask, instantiateSubTask.

Proposal at: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/201002/msg00005.html

The TC approved the proposal with the following changes:

Dieter Koenig: Three minor suggestions:
(1) Rename getParentTasksIdentifier to getParentTaskIdentifier (as in the spec -- this is probably just a typo)
(2) Prefix some of the WSDL part names like subTask or parentTask (four places) -- in the case of getParentTaskIdentifier, this avoids interpreting the taskIdentifier as an in/out parameter (the task ids are different in request and response)
(3) The result element of getParentTaskIdentifier and getParentTask should be optional (minOccurs="0") -- not every task has a parent

Dieter Koenig: (1) is just about the request/response elements, the wsdl message and operation seem to be ok




[BP-133] Result Aggregation Default Behavior

Created: 25/Jan/10  Updated: 03/Mar/10

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

PR Comment

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Reported By: Dieter König

Target: WS-HumanTask 1.1 CD 06 and PRD 01, section 4.8 Completion Behavior and section 7.2 XPath Extension Functions

Description:

WS-HT 1.1 section 4.8.2.1 describes the result construction from parallel subtasks using declarative result aggregation. In addition, section 7.2 defines XPath aggregation functions helping a WS-HT processor in performing this result aggregation. Section 7.2 also describes the returned value of an XPath function invoked with an empty node set.

This description is incomplete.

In addition to describing the XPath function results, it must be described what the parent task's output field will look like (in XML terms -- "NaN"
is just an XPath convention). This description has to consider a number of cases for the XML Schema definition of the aggregated XML value:
      Aggregation of XML elements vs. attributes
      Mandatory (minOccurs="1") vs. optional (minOccurs="0") elements
      Mandatory (use="required") vs. optional (use="optional") attributes
      Default value specified (default="...") vs. unspecified
      Null value allowed (nillable="true") vs. disallowed
      (nillable="false")

Finally, by default, the numeric XPath functions themselves should all return "NaN" for consistency reasons.

Proposal:

Between the end of the normative text in section 4.8.2.1 and the examples at the end of the section, add the following text.

<new_paragraph>

If a declarative result aggregation is applied, it is still possible that no values can be provided for the aggregation of a particular output field, for example, if no subtask has set a value to an optional field (by omission or by an explicit nil value).

In this case, the following rules determine how the aggregated output field of the parent task is set.
   (1) -- If the result value is optional (element defined with
   minOccurs="0" or attribute defined with use="optional") then the
   corresponding element or attribute in the parent task output MUST be
   omitted.
   (2) -- If (1) does not apply and a default value is provided (element or
   attribute with default="{value}") then the parent task output element or
   attribute MUST be explicitly set to this default value.
   (3) -- If (1)-(2) do not apply and the result value is a nillable
   element (element defined with nillable="true") then the parent task
   output element MUST be set to a nil value (<a xsi:nil="true"/>).
   (4) -- If (1)-(3) do not apply, that is, the result is mandatory
   (element defined with minOccurs="1" or attribute defined with
   use="required") but a value cannot be supplied, then a standard fault
   htd:aggregationFailure MUST be thrown to indicate a non-recoverable
   error.

</new_paragraph>

Finally, at the end of section 7.2, change the default behavior of the "sum" XPath function from:
   "Returns the sum value of all number nodes - returns 0 for an empty
   node-set"
to:
   "Returns the sum value of all number nodes - returns NaN for an empty
   node-set"





[BP-132] Mismatch between the specification text and XSD/XSD snippets in the document

Created: 28/Oct/09  Updated: 01/Nov/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target: WS-HumanTask 1.1 CD 05 revision 10
 
Issue Description:
Elements "peopleAssignment" and "presentationElements" are defined as optional elements in section 4.1 (Overall Syntax) and ws-humantask.xsd.
Section 4.2 says that these elements are mandatory.
Remove the inconsistency between the specification text and the XSD.
 
Proposal:
Make the following changes in section 4.2, lines 1374 and 1392:
 
Include
 
"The element is optional."
 
Instead of
 
"The element is mandatory."
 




[BP-131] Need Conformance Sections for HT and B4P specs

Created: 27/Oct/09  Updated: 01/Nov/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by Dieter Koenig

Target:

WS-HumanTask 1.1 CD 05 revision 10, section 12 Conformance BPEL4People 1.1 CD 05 revision 3, section 8 Conformance

Description:

Both sections are empty.

Proposal:

Add the following content:

WS-HT, 12. Conformance

The XML schema pointed to by the RDDL document at the namespace URI, defined by this specification, are considered to be authoritative and take precedence over the XML schema defined in the appendix of this document.

There are four conformance targets defined as part of this specification: a WS-HumanTask Definition, a WS-HumanTask Processor, a WS-HumanTask Parent and a WS-HumanTask Client (see section 2.3). In order to claim conformance with WS-HumanTask 1.1, they MUST comply with all normative statements in this specification (for each conformance target, respectively), notably all MUST statements have to be implemented.

B4P, 8. Conformance

The XML schema pointed to by the RDDL document at the namespace URI, defined by this specification, are considered to be authoritative and take precedence over the XML schema defined in the appendix of this document.

There are four conformance targets defined as part of this specification: a BPEL4People Definition, a BPEL4People Processor, a WS-HumanTask Definition and a WS-HumanTask Processor (see section 2.3). In order to claim conformance with BPEL4People 1.1, they MUST comply with all normative statements in the BPEL4People and the WS-HumanTask specification (for each conformance target, respectively), notably all MUST statements have to be implemented.

 

 Comments 

 

 

Comment by Luc Clement [ 28/Oct/09 10:46 AM ]

Proposal accepted with the following amendments:

Phil: Proposal: they => the implementations of these conformance targets
Sean Gabriel: Discussion -> to address ambiguity of "they", need to clearly define which conformance targets are being referenced
Dave Ings: Proposal: amenda Dieter's draft by replaceing the word "they" with "the conformance targets" and deleting the text in parenthesis.
Sean Gabriel: Motion to accept BP-131 (changing "they" with Dave's proposal, applied to each spec)




[BP-130] Name of startDeadline/completionDeadline must be unique

Created: 26/Oct/09  Updated: 01/Nov/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Ivana Trickovic

Target:
 
WS-HumanTask 1.1 CD 05 revision 10
 
Issue Description:
Proposal for issue BP-127 introduced operations that use name of startDeadline/completionDeadline as input parameter. The specification must mandate that the name of startDeadline and completionDeadline are unique among the names of all deadlines specified within a task definition.
 
 
Proposal:
 
In section 4.9 make the following change: Add sentence
 
"The value of the name attribute MUST be unique for all deadline specifications within a task definition."
 
at the end of the following paragraph:
 
"The element <deadlines> is used to include the definition of all deadlines within the task definition. It is optional. If present then the WS-HumanTask Definition MUST specify at least one deadline. Deadlines defined in ad-hoc sub tasks created at runtime MUST NOT contradict the deadlines of their parent task."
 




[BP-129] XPath Context Undefined

Created: 19/Oct/09  Updated: 28/Oct/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask 1.1 CD 05 revision 9
BPEL4People 1.1 CD 05 revision 2

Description:

When XPath 1.0 is used as an Expression Language in BPEL4People or WS-HumanTask language elements then the XPath context is undefined.

The following proposal is inspired by WS-BPEL 2.0 section 8.2.4 (
http://docs.oasis-open.org/wsbpel/2.0/OS/wsbpel-v2.0-OS.html#_Ref130811481
).

Proposal:

In the WS-HumanTask specification, add a new section:
========================================
2.6 Default use of XPath 1.0 as an Expression Language The XPath 1.0 specification [XPATH 1.0] defines the context in which an XPath expression is evaluated. When XPath 1.0 is used as an Expression Language in WS-HumanTask language elements then the XPath context is initialized as follows:
-- Context node: none
-- Context position: none
-- Context size: none
-- Variable bindings: none
-- Function library: Core XPath 1.0 and WS-HumanTask functions MUST be available and processor-specific functions MAY be available
-- Namespace declaration: all in-scope namespace declarations from the enclosing element

Note that XPath 1.0 explicitly requires that any element or attribute used in an XPath expression that does not have a namespace prefix must be treated as being namespace unqualified. As a result, even if there is a default namespace defined on the enclosing element, the default namespace will not be applied.
========================================

In the BPEL4People 1.1 specification, add a new section:
========================================
2.6 Default use of XPath 1.0 as an Expression Language The XPath 1.0 specification [XPATH 1.0] defines the context in which an XPath expression is evaluated. When XPath 1.0 is used as an Expression Language in BPEL4People or inlined WS-HumanTask language elements then the XPath context is initialized as follows:
-- Context node: none
-- Context position: none
-- Context size: none
-- Variable bindings: all WS-BPEL variables visible to the enclosing element as defined by the WS-BPEL scope rules
-- Function library: Core XPath 1.0, WS-BPEL, BPEL4People and WS-HumanTask functions MUST be available and processor-specific functions MAY be available
-- Namespace declaration: all in-scope namespace declarations from the enclosing element

Note that XPath 1.0 explicitly requires that any element or attribute used in an XPath expression that does not have a namespace prefix must be treated as being namespace unqualified. As a result, even if there is a default namespace defined on the enclosing element, the default namespace will not be applied.
========================================

 

 Comments 

 

 

Comment by Dave Ings [ 21/Oct/09 10:32 AM ]

Opened as per 10/21 TC.




[BP-128] CreateLeanTask Interaction Styles

Created: 10/Oct/09  Updated: 28/Oct/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

File Attachments:

Description: Microsoft WordLean Task Interactions + DK.doc     Description: Filews-humantask-leantask-api.wsdl    

 

 Description 

 

 

Submitted by: Dieter Koenig

Refer to: http://lists.oasis-open.org/archives/bpel4people/200910/msg00009.html

Target:

WS-HumanTask 1.1 CD 05 revision 8
 + Issue 111 resolution attached to
http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200910/msg00003.html
 + Issue 111 resolution amendment within TC call minutes in
http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200910/pdf00000.pdf

Description:

The amended issue 111 resolution describes "createLeanTask" as a one-way
operation. This operation definition must be revised such that the invoked
lean task's output can be returned.

Proposal:

In the same way as an operation implemented by a standard WS-HT task, the
"createLeanTask" operation should support two interaction styles:
(1) A request-response operation returning the lean task's output -- note
that this is a long-running operation that can only be invoked via
asynchronous protocols.
(2) A one-way operation that also accepts an EPR which is used for a
callback in order to deliver the lean task's output.

Revise the CreateLeanTask operation definition as shown in the attached
document "Lean Task Interactions + DK.doc" (see change tracking). Note that
this proposal document also contains additional formatting and a minor
simplification of lean task operation names.

Introduce a new WS-HT XML namespace on page 2 of the WS-HumanTask
specification:

      htlt -
      http://docs.oasis-open.org/ns/bpel4people/ws-humantask/leantask/api/200803


and a new WSDL artifact defining the lean task operations (see attached
file "ws-humantask-leantask-api.wsdl").

(See attached file: Lean Task Interactions + DK.doc)(See attached file:
ws-humantask-leantask-api.wsdl)

 

 Comments 

 

 

Comment by Dieter Koenig [ 11/Oct/09 03:49 AM ]

Revision of WS-HT 1.1 Chapter 9 (Lean Tasks)

Comment by Dieter Koenig [ 11/Oct/09 03:50 AM ]

New WS-HT 1.1 WSDL artifact (Lean Task API Operations)

Comment by Dave Ings [ 21/Oct/09 10:20 AM ]

Opened as per 10/21 TC.




[BP-127] WS-HT API's required to set task deadlines

Created: 07/Oct/09  Updated: 28/Oct/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Reported by: Ralf Mueller

Target:

Latest WS-HT spec

Description:

As part of the the discussion on BP-104, a new set of WS-HT API's needs to be added to
set a task deadlines duration or deadline expression of the startDeadline and completionDeadline

Proposal:

1. Modify type "tDeadline" in ws-humantask.xsd to include a name attribute
    for deadlines. This is to identify a deadline out of the potential many deadlines
   a task might specify
  <xsd:complexType name="tDeadline">
    <xsd:complexContent>
      <xsd:extension base="tExtensibleElements">
        <xsd:sequence>
          <xsd:choice>
            <xsd:element name="for" type="tDuration-expr" />
            <xsd:element name="until" type="tDeadline-expr" />
          </xsd:choice>
          <xsd:element name="escalation" type="tEscalation" minOccurs="0" maxOccurs="unbounded" />
        </xsd:sequence>
        <xsd:attribute name="name" type="xsd:NCName" use="required"/>
      </xsd:extension>
    </xsd:complexContent>
  </xsd:complexType>
  
2. WS-HT Chapter 7.1.1., Participant Operations
Add the following API's:

setTaskStartDeadlineExpression
In:
  task identifier
  deadline name
  deadline expression
Out:
  void

setTaskStartDurationExpression
In:
  task identified
  deadline name
  duration expression
Out:
  void

setTaskCompletionDeadlineExpression
In:
  task identifier
  deadline name
  deadline expression
Out:
  void

setTaskCompletionDurationExpression
In:
  task identifier
  deadline name
  duration expression
Out:
  void
  
Operations above should support Batch Processing, the Pre-State is CreatedReady, Reserved, In Progress and
there is no state transition as part of above operations

Please note that above API's simply set the duration or deadline expressions of a task deadline. As an alternative
we could have the API overwrite the entire startDeadline, completionDeadline definition (including the deadline
escalation definitions). Personally, i wouldn't go that route though.

 

 Comments 

 

 

Comment by Dave Ings [ 21/Oct/09 10:14 AM ]

Opened as per 10/21 TC.




[BP-126] NEW ISSUE: RDDL document creation for declared namespaces

Created: 19/Sep/09  Updated: 28/Nov/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Task

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dave Ings

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target:
 
  New document(s)
 
Description:
 
When we publish to docs.oasis-open.org (required as part of public review) we need to provide a RDDL document. A RDDL document - if an http namespace is declared - contains meta-data about the specifications, and our namespace URLs need to point to such a documents. An example is WS-BPEL 2.0:http://docs.oasis-open.org/wsbpel/2.0/wsbpel-v20-rddl.html
 
Proposal:
 
  Using the template at http://docs.oasis-open.org/templates/rddl.html ,edit specific versions for our specs. It is not clear to me whether we need to have one rddl per
  spec, so this is up for discussion/clarification.

 

 Comments 

 

 

Comment by Dave Ings [ 23/Sep/09 10:23 AM ]

Opened as per 9/23 TC.

Comment by Dave Ings [ 04/Nov/09 10:30 AM ]

Resolved as per 11/04 TC minutes.




[BP-125] OrganizationalEntity Cannot Contain Users AND Groups

Created: 10/Aug/09  Updated: 29/Sep/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask 1.1 CD 05 revision 03, Section 3.4.4 Data Type for
Organizational Entities and "ws-humantask-types.xsd". Also affected are
BPEL4People 1.1 CD 05 and example XML artifacts.

Description:

Section 3.4.4 defines the tOrganizationalEntity data type. It is used in
literal people assignments and API operations like forward or nominate.

Queries used to resolve logical people groups may return users or groups.
Moreover, they may also return mixed sets of users and groups. Likewise,
other occurrences of tOrganizationalEntity may also require support for
such mixed sets.

The choice allows an element of this type to contain EITHER a list of zero
or more users OR a list of zero or more groups, but not users AND groups.

<xsd:element name="organizationalEntity" type="tOrganizationalEntity"/>
<>xsd:complexType name="tOrganizationalEntity">
  <xsd:choice
    <xsd:element ref:="users"/>
    <xsd:element ref:="groups"/>
  </xsd:choice>
</xsd:complexType>

<xsd:element name="user" type="tUser"/>
<>xsd:simpleType name="tUser">
  <xsd:restriction base="xsd:string"/>
</xsd:simpleType>

<xsd:element name="users" type="tUserlist"/>
<>xsd:complexType name="tUserlist">
  <xsd:sequence>
    <xsd:element ref:="user" minOccurs="0" maxOccurs="unbounded"/>
  </xsd:sequence>
</xsd:complexType>

<xsd:element name="group" type="tGroup"/>
<>xsd:simpleType name="tGroup">
  <xsd:restriction base="xsd:string"/>
</xsd:simpleType>

<xsd:element name="groups" type="tGrouplist"/>
<>xsd:complexType name="tGrouplist">
  <xsd:sequence>
    <xsd:element ref:="group" minOccurs="0" maxOccurs="unbounded"/>
  </xsd:sequence>
</xsd:complexType>


Proposal:

Also allow the presence of users AND groups in an element of type
tOrganizationalEntity.

(A) Change the first paragraph in 3.4.4 from:
"The following XML schema definition describes the format of the data that
is returned at runtime when evaluating a logical people group. The result
can contain either a list of users or a list of groups. The latter is used
to defer the resolution of one or more groups of people to a later point,
such as when the user accesses a task list."

to:
"The following XML schema definition describes the format of the data that
is returned at runtime when evaluating a logical people group. The result
can contain a list of users or groups. A group is used to defer the
resolution of a group of people to a later point, such as when the user
accesses a task list."

(B) Change the XML Schema type definitions as follows (also in
"ws-humantask-types.xsd"):

<xsd:element name="organizationalEntity" type="tOrganizationalEntity"/>
<>xsd:complexType name="tOrganizationalEntity">
  <xsd:choice minOccurs="0" maxOccurs="unbounded">
    <xsd:element name="user" type="tUser"/>
    <xsd:element name="group" type="tGroup"/>
  </xsd:choice>
</xsd:complexType>

<xsd:element name="user" type="tUser"/>
<>xsd:simpleType name="tUser">
  <xsd:restriction base="xsd:string"/>
</xsd:simpleType>

<xsd:element name="group" type="tGroup"/>
<>xsd:simpleType name="tGroup">
  <xsd:restriction base="xsd:string"/>
</xsd:simpleType>

(C) Make all operation signatures referencing the existing list types
tUsers and tGroups consistent with the above definition.

(D) Make all examples using elements of type tOrganizationalEntity, tUsers
and tGroups in WS-HT, B4P and example XML artifacts consistent with the
above definition.

 

 Comments 

 

 

Comment by Dave Ings [ 22/Sep/09 08:15 AM ]

Opened as per 9/16 TC.

Comment by Dave Ings [ 22/Sep/09 08:17 AM ]

As per 9/16 TC

Comment by Dave Ings [ 22/Sep/09 08:18 AM ]

Resolved as per 9/16 TC minutes.




[BP-124] Location Attribute Mandatory in Result Aggregation

Created: 05/Aug/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Major

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target: WS-HumanTask CD 05 revision 02, Section 4.7.2 Declarative Result Aggregation

Description:

Section 4.7.2 defines the <aggregate> element used to combine output from multiple subtasks.
<htd:aggregate location="query" condition="bool-expr"? function= "function-call"/>

If a task interface (or an individual WSDL part, see also issue BP-123) only has one simple-typed element then the location query can be optional.
The aggregation function is then applied to this single simple-typed element. Example: an approval step which just returns a boolean value.

Proposal:

In section 4.7.2, make the @location attribute optional within the <aggregate> element:
<htd:aggregate location="query"? condition="bool-expr"? function= "function-call"/> (question mark added to @location)

In the explanation, change from:
"The mandatory location attribute MUST contain ..."
to:
"The optional location attribute MUST contain ..."




[BP-123] WSDL Part Name Missing in Result Aggregation

Created: 05/Aug/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask CD 05 revision 02, Section 4.7.2 Declarative Result Aggregation

Description:

Section 4.7.2 defines the <aggregate> element used to combine output from multiple subtasks:
<htd:aggregate location="query" condition="bool-expr"? function= "function-call"/>

An interface of a task (and its subtasks) may be defined using a WSDL message with multiple WSDL parts. In this case, the @location attribute is not sufficient to uniquely identify the source and target of the aggregation function. It is unclear to which WSDL part the location query is applied.

Proposal:

In section 4.7.2, add an optional @part attribute to the <aggregate>
element:
<htd:aggregate part="NCName"? location="query" condition="bool-expr"?
function="function-call"/>

with the following explanation:
"The optional part attribute MUST contain the name of a WSDL part. The part attribute MUST be specified when the task interface is defined using a WSDL message with more than one WSDL part."




[BP-122] Task Definition Structure

Created: 03/Aug/09  Updated: 07/Oct/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

No Action

Votes:

0

 

 

 Description 

 

 

Submitted by:

Target: WS-HumanTask CD 05 revision 02, Section 4 Human Tasks and Section 5 Lean
Tasks; ws-humantask.xsd

Description:

The definition of composite tasks requires three elements for each
composition nesting level.

<task name="...">
  <composition type="..." activationPattern="...">?
    <subtask name="...">+
      <task name="...">>
        <composition type="..." activationPattern="...">?
          <subtask name="...">+
            ...

This is an unnecessary syntax complexity -- one element nesting level can
be saved for each composition nesting level.

Moreover, both subtask and task have a @name attribute -- this is confusing
when referencing the name in completion conditions or result aggregations.

Finally, the overriding elements in the <localTask> element create an
unnecessary complexity within a set of task definitions.


Proposal:

   (See attached file: Task Definition Structure.doc)(See attached file:
   ws-humantask - work in progress.xsd)

Use one composition element with @compositionType and @activationPattern
attributes. Add a @name attribute to the <localTask> element such that each
child of <subtasks> has a name. Remove the overriding elements from
<localTask>.

<task name="...">
  <subtasks compositionType="..." activationPattern="...">?
    <task name="...">+
      <subtasks compositionType="..." activationPattern="...">?
        ...

Submission email: http://lists.oasis-open.org/archives/bpel4people/200908/msg00000.html.

See the attachments for more details. Note that the XML Schema file already
anticipates other issue resolutions such as BP-97, BP-102 (which are not
important for this issue discussion).

 

 Comments 

 

 

Comment by Luc Clement [ 07/Oct/09 10:37 AM ]

We decided to close with no action.




[BP-121] Optional PeopleAssignments and PresentationElements

Created: 29/Jul/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask CD 05 revision 02, section 4.1 Overall Syntax

Description:

In the syntax definition for human tasks, the two elements
<peopleAssignments> and <presentationElements> are mandatory.

For composite tasks, routing patterns, and lean tasks, they should not be
required.

Proposal:

Make both elements optional, that is, add a '?' to <peopleAssignments> and
<presentationElements> within the <task> syntax in 4.1.




[BP-120] Missing (Forbidden) Composition Element in Lean Tasks

Created: 29/Jul/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask CD 05 revision 02, section 5.1 Lean Tasks

Description:

The new element <htd:composition> can be used in a human task definition in
order to define a composite task. However, it should be explicitly
forbidden in a lean task definition.

Proposal:

Add the element

  <htd:composition>
    ...
  </htd:composition>

at the end of the syntax definition for the <leanTask> element in section
5.1, but show it using a *** strike through *** style in order to clarify
that it is forbidden in lean tasks.




[BP-119] Missing Composition Element

Created: 29/Jul/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Major

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask CD 05 revision 02, section 4.1 Overall Syntax

Description:

The new element <htd:composition> is missing in the syntax of the human
task definition.

Proposal:

Add the element

  <htd:composition>?
    ...
  </htd:composition>

at the end of the syntax definition for the <task> element in section 4.1.




[BP-118] Usage of Term "Task Type" in WS-HT

Created: 22/Jul/09  Updated: 24/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ravi Rangaswamy

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter König

Target: WS-HumanTask CD 05 revision 02

Description:

The term "task type" is used inconsistently -- sometimes in the sense of "task definition" (as opposed to "task instance") and sometimes in the sense of "task vs. notification" (e.g. in the API).

Proposal: http://lists.oasis-open.org/archives/bpel4people/200908/msg00004.html




[BP-117] Usage of Term "People Activity" in WS-HT

Created: 01/Jul/09  Updated: 15/Jul/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask CD 04 revision 04, section 3.2 Composite Tasks and Sub Tasks, second paragraph

Description:

This paragraph has a sentence "Therefore the existence of sub tasks usually won't be visible on people activity level".

There is no concept of people activities in WS-HT and there must not be a reference to BPEL4People.

Proposal:

Drop this sentence.

 

 Comments 

 

 

Comment by Dave Ings [ 08/Jul/09 03:52 PM ]

Opened at 7/8 TC.

Comment by Dave Ings [ 08/Jul/09 03:53 PM ]

Proposed resolution accepted at 7/8 TC.




[BP-116] Missing Pre-State / Post-State Specifications for Administrative Operations

Created: 29/Jun/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

File Attachments:

Description: Microsoft WordBP-116 - ws-humantask-1.1-spec.doc     Description: Microsoft WordBP-116 rev1 - ws-humantask-1.1-spec.doc    

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask CD 04 revision 03, section 7.1.4 Operation Authorizations

Description:

Section 7.1.4 introduces client operations for administrative purposes.
Unlike section 7.1.1, which specifies the general-purpose client operations, the table in 7.1.4 has no columns defining the task pre-states in which an operation may be called and the task post-state after the operation is performed.

Proposal:.

Add the following paragraph (copied from 7.1.1) before the table in 7.1.4:

<newParagraph>
If the task is in a predefined state listed as valid pre-state before the
operation is invoked then, upon successful completion, the task MUST be in
the post state defined for the operation. If the task is in a predefined
state that is not listed as valid pre-state before the operation is invoked
then the operation MUST be rejected and MUST NOT cause a task state
transition.
</newParagraph>

In addition, add two columns "Pre-State" and "Post-State" to the table in
7.1.4, with the following content:

+---------------------+ ... +-------------+--------------+
| Operation Name | ... | Pre-State | Post-State |
+---------------------+ ... +-------------+--------------+
| activate | ... | Created | Ready |
+---------------------+ ... +-------------+--------------+
| nominate | ... | Created | Ready |
| | ... | | Reserved |
+---------------------+ ... +-------------+--------------+
| setGenericHumanRole | ... | Created | (no state |
| | ... | Ready | transition) |
| | ... | Reserved | |
| | ... | InProgress | |
+---------------------+ ... +-------------+--------------+

 

 Comments 

 

 

Comment by Dieter Koenig [ 04/Jul/09 09:20 AM ]

The file attachment shows the proposed two new columns for the table that defines administrative operations.

Comment by Dieter Koenig [ 22/Jul/09 11:12 AM ]

Revised proposal - setGenericHumanRole can now be called in all non-final states (I added the Suspended/* states)




[BP-115] Missing Authorization Specifications for Batch Operations

Created: 29/Jun/09  Updated: 18/Jul/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Bug

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask CD 04 revision 03, section 7.1.5 Operation Authorizations

Description:

The resolution of BP-69 introduced a number of batch operations (see section 7.1.1):
 - batchClaim
 - batchStart
 - batchStop
 - batchRelease
 - batchSuspend
 - batchSuspendUntil
 - batchResume
 - batchComplete
 - batchRemove
 - batchFail
 - batchSetPriority
 - batchSkip
 - batchForward
 - batchDelegate
 - batchActivate
 - batchNominate
 - batchSetGenericHumanRole

The specification of the corrresponding operation authorizations are missing.

Proposal:

Batch operations interact with multiple tasks simultaneously, and the authorization for each interaction has to be decided individually.
Consequently, add the following paragraph to section 7.1.5 Operation Authorizations, right before the table that defines the operation-specific
authorizations:

<newParagraph>
All batch operations (operations with a name prefix "batch") may be invoked by any caller; no specific authorization is required. Missing authorizations for operations on individual tasks result in a report entry in the batch operation's response message.
</newParagraph>

In addition, add a generic entry to the table (operation name = "batch*", all entries = "+").




[BP-114] Missing Spec Text in Section 4.6 Elements for Composite Tasks

Created: 26/Jun/09  Updated: 02/Sep/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:

WS-HumanTask CD 04 revision 04, section 4.6 Elements for Composite Tasks

Description:

The section has one syntax definition element but no text -- introduction text must be added, explaining the introduced syntax.

Proposal: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200908/msg00025.html with the following amendment:

Change
- Original text: creationPattern: This optional attribute specifies the way how sub-tasks are created. If the value is set to manual the task client triggers creation of enclosed sub-tasks. Otherwise, they are automatically created at the time the composite task itself turns into status inProgress. The default value for this attribute is manual.

To:
- Changed text: instantiationPattern: This optional attribute specifies the way how sub-tasks are instantiated. If the value is set to manual the task client triggers instantiation of enclosed sub-tasks. Otherwise, they are automatically instantiated at the time the composite task itself turns into status inProgress. The default value for this attribute is manual.

 

 Comments 

 

 

Comment by Dave Ings [ 08/Jul/09 04:07 PM ]

Resolved due to fat fingers. Will be reopened.




[BP-113] Missing Authorization Specifications for new Operations

Created: 25/Jun/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by Dieter Koenig

Target:

WS-HumanTask CD 04 revision 03

Description:

The resolution of BP-28 introduced a number of new operations (see section 7.1.1):
- getSubtasks
- getSubtaskIdentifiers
- hasSubtasks
- getParentTask
- getParentTaskIdentifier
- isSubtask

The corresponding operation authorizations (see section 7.1.5) are missing..

Proposal:

Posted with http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200907/msg00017.html




[BP-112] Aggregation Functions Invoked with Empty Node-Set

Created: 24/Jun/09  Updated: 18/Jul/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter Koenig

Target:
   WS-HumanTask 1.1 CD 04 rev 2, section 6.2, second table: XPath functions
   for Result Aggregation.

Description:
   The aggregation function specifications do not describe the return value
   for the case where an empty node-set is passed.

Proposal:
   See inlined text in the table of aggregation functions below. As the
   color code may be lost during mail transmission, I also attached a
   document showing the changes.

   (See file: XPath Extension Functions.doc attached to email http://lists.oasis-open.org/archives/bpel4people/200906/msg00071.html)
                                                                                       
                               | |
         concat | Returns the | In
                               | concatenation of all |
                               | string nodes - |
                               | returns an empty | • node-set of string
                               | string for an empty | nodes
                               | node-set |
       ------------------------+-----------------------+------------------------------
                               | |
         concatWithDelimiter | Returns the | In
                               | concatenation of all |
                               | string nodes, |
                               | separated by the | • node-set of string
                               | specified delimiter | nodes
                               | string - returns an |
                               | empty string for an |
                               | empty node-set | • delimiter string
       ------------------------+-----------------------+------------------------------
                               | |
         leastFrequentOccurenc | Returns the least | In
         e | frequently occurring |
                               | string value within |
                               | all string nodes, or | • node-set of string
                               | an empty string in | nodes
                               | case of a tie or for |
                               | an empty node-set |
       ------------------------+-----------------------+------------------------------
                               | |
         mostFrequentOccurence | Returns the most | In
                               | frequently occurring |
                               | string value within |
                               | all string nodes, or | • node-set of string
                               | an empty string in | nodes
                               | case of a tie or for |
                               | an empty node-set |
       ------------------------+-----------------------+------------------------------
                               | |
         voteOnString | Returns the most | In
                               | frequently occurring |
                               | string value if its |
                               | occurrence is above | • node-set of string
                               | the specified | nodes
                               | percentage and there |
                               | is no tie, or an |
                               | empty string | • percentage value in
                               | otherwise (including | the range from 0 to
                               | an empty node-set) | 100
       ------------------------+-----------------------+------------------------------
                               | |
         and | Returns the | In
                               | conjunction of all |
                               | boolean nodes - |
                               | returns false for an | • node-set of boolean
                               | empty node-set | nodes
       ------------------------+-----------------------+------------------------------
                               | |
         or | Returns the | In
                               | disjunction of all |
                               | boolean nodes - |
                               | returns false for an | • node-set of boolean
                               | empty node-set | nodes
       ------------------------+-----------------------+------------------------------
                               | |
         vote | Returns the most | In
                               | frequently occurring |
                               | boolean value if its |
                               | occurrence is above | • node-set of boolean
                               | the specified | nodes
                               | percentage, or false |
                               | otherwise (including |
                               | an empty node-set) | • percentage value in
                               | | the range from 0 to
                               | | 100
       ------------------------+-----------------------+------------------------------
                               | |
         avg | Returns the average | In
                               | value of all number |
                               | nodes - returns NaN |
                               | for an empty | • node-set of number
                               | node-set | nodes
       ------------------------+-----------------------+------------------------------
                               | |
         max | Returns the maximum | In
                               | value of all number |
                               | nodes - returns NaN |
                               | for an empty | • node-set of number
                               | node-set | nodes
       ------------------------+-----------------------+------------------------------
                               | |
         min | Returns the minimum | In
                               | value of all number |
                               | nodes - returns NaN |
                               | for an empty | • node-set of number
                               | node-set | nodes
       ------------------------+-----------------------+------------------------------
                               | |
         sum | Returns the sum | In
                               | value of all number |
                               | nodes - returns 0 |
                               | for for an empty | • node-set of number
                               | node-set | nodes
                                                                                       


 

 Comments 

 

 

Comment by Dave Ings [ 08/Jul/09 04:08 PM ]

Resolved at 7/8 TC.




[BP-111] Addition Register, Unregister and List to support interop of Task Parent and Task Processor

Created: 19/Jun/09  Updated: 14/Oct/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Fixed

Votes:

0

 

Issue Links:

Depends on

is depended on by

BP-109

Lean Task - addition of lean tasks in...

Applied

 

 Description 

 

 

Submitted by: Alex Malek

Target: HT

Description: Lean Tasks - should we add Register, Unregister and List to support full interop of Task Parents and Task Processors

Proposal: http://lists.oasis-open.org/archives/bpel4people/200910/msg00003.html with the following amendment: 9.2.5 should have the three references to the word 'protocol' removed and to remove the CreateLeanTaskResponse and all references to it.

 

 Comments 

 

 

Comment by Dave Ings [ 08/Jul/09 04:16 PM ]

Opened at 7/8 TC.

Comment by Dave Ings [ 06/Oct/09 03:29 PM ]

Was closed in error. I have reopened it.

Comment by Luc Clement [ 07/Oct/09 10:36 AM ]

Motion to resolve Issue 111 with the text, except 9.2.5 should have the three references to the word 'protocol' removed and to remove the CreateLeanTaskResponse and all references to it.




[BP-110] Explaining the restriction of interface in lean tasks schema

Created: 19/Jun/09  Updated: 08/Jul/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

 

 Description 

 

 

Submitted by: Alex Malek

Target: HT

Description: Lean Task - how should we explain the restriction of interface in lean tasks schema

 

 Comments 

 

 

Comment by Dave Ings [ 08/Jul/09 03:52 PM ]

Closed no action at 7/8 TC.




[BP-109] Lean Task - addition of lean tasks in the arch overview

Created: 19/Jun/09  Updated: 29/Sep/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

depends on

BP-111

Addition Register, Unregister and Lis...

Applied

 

 Description 

 

 

Submitted by: Alex Malek

Target: HT

Description: addition of lean tasks in the arch overview

Accepted Proposal: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200909/msg00028.html

 

 Comments 

 

 

Comment by Dave Ings [ 08/Jul/09 03:51 PM ]

Opened at 7/8 TC. Alireza and Frank to assist Alex.

Comment by Dave Ings [ 05/Aug/09 03:44 PM ]

Dependency added as per 8/5 TC.




[BP-108] Ambiguity: routing pattern / parallel composite task creation

Created: 19/Jun/09  Updated: 12/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ravi Rangaswamy

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

depends on

BP-92

HT section 3 need new text defining t...

Applied

 

 Description 

 

 

Submitted by: Phil Allen

Target: WS-HT

Issue Description:

The spec is ambiguous about which of the options is the correct way to implement the specification, and can lead to very different, incompatible implementations.

Given the following task element with both a 3-person parallel routing pattern and a 3-person parallel composite task:
<task>
    <peopleAssignments>
         <potentialOwners>
              <parallel type="all">
                    <htt:users>
                     <htt:user>A</htt:user>
                     <htt:user>B</htt:user>
                     <htt:user>C</htt:user>
                    </htt:users>
             </parallel>
          </potentialOwners>
     </peopleAssignments>
     <composition type="parallel" activationProperties="automatic">
         <subTask name="a" multiInstance="sequential">
             <task>
                 <peopleAssignments>
                    <potentialOwners>
                       <parallel type="all">
                          <htt:users>
                          <htt:user>D</htt:user>
                          <htt:user>E</htt:user>
                          <htt:user>F</htt:user>
                         </htt:users>
                      </parallel>
                   </potentialOwners>
               </peopleAssignments>
            </task>
          </subTask>
      </composition>
 
In this case, how many tasks are created?

Options:
1) 6. Three for the first parallel, three for the second parallel, all in parallel with each other.
2) 12. Three for the first parallel, and three for each of these members of the routing patterns parallel.
3) 0. It is an error to use both.
4) 3. Three for the first parallel, and the composition is ignored because of the routing pattern.
5) 4. One task for the root task element (no routing pattern, potential owners = A or B or C) and three for the composition.

Proposal:

 

 Comments 

 

 

Comment by Dave Ings [ 08/Jul/09 03:47 PM ]

Opened at 7/8 TC. Oliver to assist because of link to BP-92.

Comment by Ravi Rangaswamy [ 21/Jul/09 08:40 PM ]

Proposal after discussion with Oliver and Ralf:

When composition is defined on a task, the composition is applied for each of the potential owners defined in the task's people assignment.

So, in the example in the BP-108, there would be 13 tasks created:

1) 1 wrapper task for overall task (no actual owners on this task)
2) 3 sub tasks - one sub task for each A, B and C
3) 9 sub tasks - Three task for D, E and F for each of the sub tasks created in step 2

We also propose that we create a new sub section in section 4 detailing the relationship between composite tasks and routing patterns (BP-92) and also capture this resolution in that sub section.




[BP-107] HT 6.1.1 clarify complete operation w.r.t the sequential pattern

Created: 18/Jun/09  Updated: 08/Jul/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Martin Chapman

Resolution:

No Action

Votes:

0

 

Issue Links:

Depends on

depends on

BP-85

Sequential routing pattern & sub-tasks

Applied

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 6.1.1 clarify complete operation w.r.t the sequential pattern. dependent on issue-85

 

 Comments 

 

 

Comment by Dave Ings [ 08/Jul/09 04:00 PM ]

Closed no action at 7/8 TC.




[BP-106] HT 4.9.4 add a normative statement that resume on sub-task is not allowed when parent is suspended

Created: 18/Jun/09  Updated: 29/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 4.9.4 add a normative statement that resume on sub-task is not allowed when parent is suspended link to YYY

Proposal: Mike R. moves to resolve BP-106 as proposed in jira




[BP-105] HT 4.9.8 review whole section regarding parent and subtask operation propagation

Created: 18/Jun/09  Updated: 23/Sep/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 4.9.8 review whole section regarding parent and subtask operation propagation
e.g. if a subtask faults must the parent task fault or is this configurable. We may need to refactor all of 4.9. issue YYY

Proposal approved: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200909/msg00023.html

 

 Comments 

 

 

Comment by Dave Ings [ 19/Jun/09 08:33 AM ]

Dieter will assist.




[BP-104] HT 4.8 Clarify Escalation behaviour

Created: 18/Jun/09  Updated: 29/Sep/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

File Attachments:

Description: Microsoft Wordws-humantask-1.1-spec.doc    

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 4.8 Clarify Escalation behaviour

Proposal: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200908/msg00026.html

 

 Comments 

 

 

Comment by Ralf Mueller [ 29/Jul/09 03:05 AM ]

An initial proposal for this issue is in section 4.8




[BP-103] HT 4.7.1. Should we allow the attachment of completion criteria to a task element

Created: 18/Jun/09  Updated: 14/Oct/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:
HT 4.7.1. Should we allow the attachment of completion criteria to a task element (with or without sub-tasks). If the answer is yes we need to define what happens after the completion condition evaluates to true for both automatic and manual.

 

 Comments 

 

 

Comment by Ivana Trickovic [ 15/Jul/09 09:44 AM ]

Issue proposal:

1. Allow the attachment of completion criteria to a task element, with or without sub-tasks.

The new syntax is

<htd:task name="NCName">
<htd:interface .../>?
<htd:priority...>?
<htd:peopleAssignments.../>
</htd:completionBehavior.../>?
...
</htd:task>

Element "completionBehavior" is described in section 4.7.

2. Extend the "completionBehavior" element with a "completion" attribute which may have two values: manual or automatic. If the value is set to "manual" the task MUST be completed explicitly by the actual owner as soon as the completion criteria evaluate to true. If the value is set to "automatic" that task MUST be set to complete as soon as the completion criteria evaluate to true. For routing tasks, the completion attribute MUST have value "automatic".

Comment by Ivana Trickovic [ 15/Jul/09 09:56 AM ]

Additional point:
3. For „plain" tasks, the completion criteria are evaluated while the task is in state „inProgress". Implementations MAY decide when and how often the completion criteria are evaluated.




[BP-102] HT 4.7.1 simplification of criterion syntax

Created: 18/Jun/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:
HT 4.7.1 simplification of criterion syntax (e.g. removal of 'and' operator)

Proposal: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200907/msg00037.html

 

 Comments 

 

 

Comment by Dave Ings [ 19/Jun/09 08:29 AM ]

Dieter will assist.




[BP-101] HT 4.7.1 Need to specify when completion criteria has to be evaluated

Created: 18/Jun/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 4.7.1 Need to specify when completion criteria has to be evaluated

Proposal: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200907/msg00037.html

 

 Comments 

 

 

Comment by Dave Ings [ 19/Jun/09 08:29 AM ]

Dieter will assist.




[BP-100] HT 4.6.1.2 Need to define completion behaviour for sequential

Created: 18/Jun/09  Updated: 18/Jul/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 4.6.1.2 Need to define completion behaviour for sequential

Proposal:
http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200907/msg00018.html

 

 Comments 

 

 

Comment by Luc Clement [ 15/Jul/09 10:38 AM ]

Approved as Resolved 15 July 2009




[BP-99] HT 4.6.1.1. Review the collaborate attribute and its definition

Created: 18/Jun/09  Updated: 30/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ravi Rangaswamy

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 4.6.1.1. Review the collaborate attribute and its definition (e.g. is it too broad a control), and possible remove task output text

Proposal: Proposed resolution for BP-99 is to remove the text from section 4.6.1.1 of Ravi's doc
resolving BP-28.




[BP-98] HT 4.5 sharing of comments need to be addressed

Created: 18/Jun/09  Updated: 30/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ravi Rangaswamy

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 4.5 sharing of comments need to be addressed

Proposal: Resolve issue 98 by noting that we don't need anything for comment sharing, and we should remove attachmentPropogation from section 4.5 of WS-HT




[BP-97] HT 4.5 Is Multi-instance on composite sub-task necessary?

Created: 18/Jun/09  Updated: 12/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ravi Rangaswamy

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

depends on

BP-92

HT section 3 need new text defining t...

Applied

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:

HT 4.5 Is Multi-instance on composite sub-task necessary? link to BP-92

 

 Comments 

 

 

Comment by Oliver Kieselbach (SAP) [ 20/Jul/09 09:52 AM ]

Depending on resolution BP-92: the proposed solution requires a "mutlinstance" attributes since the syntax to describe composite tasks must be expressive enough the describe also routing patterns.

Comment by Oliver Kieselbach (SAP) [ 21/Jul/09 10:20 AM ]

After discussion with Ravi and Ralf and depending on the resolution of issue BP-92
The proposal would to remove the "multiInstance" attribute




[BP-96] HT section 3.6.5 2nd paragraph move to routing patterns section 3.3

Created: 18/Jun/09  Updated: 29/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:
HT section 3.6.5 2nd paragraph move to routing patterns section 3.3

Proposal: Mike R. moves to resolve BP-96 with moving HT section 3.6.5 2nd to routing patterns section 3.3




[BP-95] HT section 3.6.5. Review 1st paragraph

Created: 18/Jun/09  Updated: 08/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:
HT section 3.6.5. Review 1st paragraph and either make more specific and/or more generic e.g. do we need new apis

Proposal: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200907/msg00051.html

Resolved as proposed at http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200907/msg00051.html except that it will remove the text that was to be changed in section 3.7.5

 

 Comments 

 

 

Comment by Dave Ings [ 19/Jun/09 08:19 AM ]

Ravi will assist.




[BP-94] HT Section 3.2.1 review the correctness of sub-task creation time

Created: 18/Jun/09  Updated: 08/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:
HT Section 3.2.1 review the correctness of sub-task creation time (e.g. is it currently too prescriptive) [probable link to 3.4.5 issue above]

Proposal:
(per http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200907/msg00023.html)

BP94-Proposed Text Change:

In case a composite task is pre-defined as such, the task model contains the definition of one or more sub-tasks. Composite tasks come with the following additional attributes:
• Composition Type (parallel | sequential)
Composite tasks with composition type "parallel" allow multiple active sub-tasks at the same time; sub-tasks are not in any order; composite tasks with composition type "sequential" only allow sequential creation of sub-tasks in the pre-defined order (a second listed sub-task must not be created before a first listed sub-task has been terminated).
• Instantiation Pattern (manual | automatic)
Composite tasks with instantiation pattern "manual" expect the "actual owner" to trigger the creation of pre-defined sub-tasks; composite tasks with instantiation pattern "automatic" are automatically created at the time the composite task itself turns into "in progress" status; (where composition type is "parallel," all pre-defined sub-tasks are created at the time the composite task turns into "in progress" status; where composition type is "sequential," at the time the composite task turns into "in progress" status, the first defined sub-task in the sequence is created; the next sub-task in the sequence is automatically created at the time its predecessor is terminated).

 

 Comments 

 

 

Comment by Dave Ings [ 19/Jun/09 08:15 AM ]

Oliver will assist.




[BP-93] HT section 3.2.1 change activation pattern to creation pattern

Created: 18/Jun/09  Updated: 29/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description: HT section 3.2.1 change activation pattern to creation pattern

Proposal: Martin motions that in section 3.2.1 and related schema that we change the term activation pattern to creation pattern, and that this be the resolution of issue BP-93.




[BP-92] HT section 3 need new text defining the relationship between composite tasks and patterns

Created: 18/Jun/09  Updated: 08/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

is depended on by

BP-108

Ambiguity: routing pattern / parallel...

Applied

is depended on by

BP-85

Sequential routing pattern & sub-tasks

Applied

is depended on by

BP-97

HT 4.5 Is Multi-instance on composite...

Applied

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:
HT section 3 need new text defining the relationship between composite tasks and patterns (Issue-85 is a sub issue of this) and validate the design principle of patterns being able to be built on top of composite tasks- this is issue XXX

 

 Comments 

 

 

Comment by Oliver Kieselbach (SAP) [ 20/Jul/09 09:50 AM ]

a proposal could look like this:

The complex people assignment used to describe Routing Patterns is a specific syntax version of Composite Tasks. It is a convenient syntax to decribe the "who" in a composite task scenario. The composite task syntax is more expressive to describe the "what" in the sense of what different subtasks are executed.
The syntax to describe complex people assignment can be mapped to the syntax to describe composite tasks. A compliant task processor may decide to describe routing patterns using the complex people assingment syntax or the composite task syntax. A composite task, including subtasks of different task types, can be desribed only using the composite task syntax.
Both syntax flavors may be used also in combination, which means that a composite task type may include a complex people assignment and that any task defining a complex people assignment may become a composite task at runtime when creating adhoc subtasks.
The runtime instantiation model and observable behavior for task instances is identical when using one or the other syntax flavor.

Comment by Oliver Kieselbach (SAP) [ 21/Jul/09 10:16 AM ]

Proposal after discussion with Ravi and Ralf:

The complex people assignment used to describe Routing Patterns is a specific syntax version of Composite Tasks. It is a convenient syntax to decribe the "who" in a composite task scenario. The composite task syntax is more expressive to describe the "what" in the sense of what different subtasks are executed.
A composite task, including subtasks of different task types, can be desribed only using the composite task syntax.
Both syntax flavors may be used also in combination, which means that a composite task type may include a complex people assignment and that any task defining a complex people assignment may become a composite task at runtime when creating adhoc subtasks.
The runtime instantiation model and observable behavior for task instances is identical when using one or the other syntax flavor.

Comment by Matthias Kloppmann (IBM) [ 29/Jul/09 09:55 AM ]

You state: "A composite task, including subtasks of different task types, can be desribed only using the composite task syntax. " which is correct. Don't we also need to say "A routing task containing a dynamic number of subtasks derived from the cardinality of the set of assigned people can be described only using the routing task syntax." ?




[BP-91] Review for correctness the default sub-task assignments

Created: 18/Jun/09  Updated: 08/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:
HT section 3.4.5 Review for correctness the default sub-task assignments (an example is the default initiator assignment for sub-tasks activated automatically).

 

 Comments 

 

 

Comment by Dave Ings [ 19/Jun/09 08:11 AM ]

Alex will assist Oliver.

Comment by Oliver Kieselbach (SAP) [ 14/Jul/09 09:50 AM ]

Proposal
1.1.1 Subtasks
Like a task, a sub task has a set of generic human roles . In case people assignment to a sub task's roles is not defined (neither in the sub task's task definition nor on composite task level (using overwrite mechanisms)) the following default assignments apply (especially valid for ad-hoc scenarios):
• Task initiator
a) Activation pattern "manual" à A WS-HumanTask Processor MAY assign the actual owner of the composite task
b) Activation pattern "automatic" à A WS-HumanTask Processor MAY assign the initiator of the composite task.
• Task stakeholders
o A WS-HumanTask Processor MAY assign the actual owner of the composite task
• Potential owners
o No default assignment (usually potential owners will explicitly be defined)
• Excluded owners
o A WS-HumanTask Processor MUST assign the excluded owners of the composite task
(This rule applies always, even though the excluded owners of a sub task may be enhanced by additional people)
• Business administrators
o A WS-HumanTask Processor MAY assign the business administrators of the composite task




[BP-90] Complete HT section 3.3, Routing Pattern intro

Created: 18/Jun/09  Updated: 24/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ravi Rangaswamy

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Martin Chapman

Target: WS-HT

Issue Description:
Complete HT section 3.3, Routing Pattern intro

Proposal at: http://lists.oasis-open.org/archives/bpel4people/200908/msg00019.html

Approved the proposal with the following amendment:

Remove the paragraph "Operations that can be performed by a parallel user
All operations like reassign, claim, etc. can be performed by an assignee working in parallel. Such operations MUST NOT alter the assignments of the other parallel branch."




[BP-89] June F2F Issues

Created: 18/Jun/09  Updated: 18/Jun/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

INVALID

Votes:

0

 

 

 Description 

 

 

The following new issues were agreed to be raised and opened at the F2F:

New Issue: Complete HT section 3.3, Routing Pattern intro

New Issue: HT section 3.4.5 Review for correctness the default sub-task assignments (an example is the default initiator assignment for sub-tasks activated automatically).

New issue: HT section 3 need new text defining the relationship between composite tasks and patterns (Issue-85 is a sub issue of this) and validate the design principle of patterns being able to be built on top of composite tasks- this is issue XXX



New issue: HT section 3.2.1 change activation pattern to creation pattern

New Issue: HT Section 3.2.1 review the correctness of sub-task creation time (e.g. is it currently too prescriptive) [probable link to 3.4.5 issue above]

New Issue: HT section 3.6.5. Review 1st paragraph and either make more specific and/or more generic e.g. do we need new apis.

New Issue: HT section 3.6.5 2nd paragraph move to routing patterns section 3.3


New Issue: HT 4.5 Is Multi-instance on composite sub-task necessary? link to XXX

New Issue: HT 4.5 sharing of comments need to be addressed

New Issue: HT 4.6.1.1. Review the collaborate attribute and its definition (e.g. is it too broad a control), and possible remove task output text


New issue: HT 4.6.1.2 Need to define completion behaviour for sequential

New Issue: HT 4.7.1 Need to specify when completion criteria has to be evaluated

New Issue: HT 4.7.1 simplification of critereon syntax (e.g. removal of 'and' operator)

New Issue: HT 4.7.1. Should we allow the attachment of completion criteria to a task element (with or without sub-tasks). If the answer is yes we need to define what happens after the completion condition evaluates to true for both automatic and manual.


New issue: HT 4.8 Clarify Escalation behaviour

New Issue: HT 4.9.8 review whole section regarding parent and subtask operation propagation
e.g. if a subtask faults must the parent task fault or is this configurable. We may need to refactor all of 4.9. issue YYY

New Issue: HT 4.9.4 add a normative statement that resume on sub-task is not allowed when parent is suspended link to YYY

New Issue: HT 6.1.1 clarify complete operation w.r.t the sequential pattern. dependant on issue-85

 

 Comments 

 

 

Comment by Luc Clement [ 18/Jun/09 10:50 AM ]

individual issues to be created from the submission




[BP-88] Enhance state diagram for composite task / routing pattern scenarios

Created: 20/May/09  Updated: 06/Aug/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Hannah Petereit
Proposed by: sub-task / routing pattern working session team

Target: WS-HT

Issue Description:

WS-HT specification, chapter 4.7 Human Task Behavior and State Transitions
Potential enhancement of state transitions to cover dependencies between composite and sub tasks.
Should we introduce a new status for composite tasks, indicating that the task itself is not processed, but sub-tasks are?

 

 Comments 

 

 

Comment by Dave Ings [ 03/Jun/09 01:06 PM ]

As per 8/3 subteam meeting, Gerhard will partner with Oliver to develop a proposal.

Comment by Gerhard Pfau [ 22/Jul/09 10:14 AM ]

Initial proposal for resolution of BP-88:

Change chapter 4.9.1 as descibed below:

Remove the following sentence at the end of the chapter:
WS-HumanTask Processor SHOULD always activate sub tasks on creation.

Add the following text at the end of the chapter:
For human tasks that have subtasks two different cases exist, with different implications:
1. Tasks with subtasks where an actual owner is desired
2. Tasks with subtasks where no actual owner is desired
The first case has the sub-case where a potential owner has been modeled on the primary task and subtasks have been modeled that are activated either manually or automatically. Another sub-case of the first case is the one where no potential owner has been modeled and thus nomination has to occur. In all cases there is an actual owner eventually and the primary task goes through the state transitions from Created to Ready to Reserved to InProgress, etc.
In the second case where no actual owner is desired the human task (the primary task) directly transitions from state Created to InProgress. Subtasks are always instantiated automatically.

Note: The state diagram has to be changed accordingly (add a transition from Created to InProgress




[BP-87] XPath functions to get output of tasks

Created: 20/May/09  Updated: 01/Jul/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

No Action

Votes:

0

 

 

 Description 

 

 

Submitted by: Hannah Petereit
Proposed by: sub-task / routing pattern working session team

Target: WS-HT

Issue Description:

WS-HT specification, chapter 6.2 XPath Extension Functions
Decide whether a generic getOutput function needs to be introduced or just a function to serve composite task / routing pattern scenarios (that allows a composite task to access output of its sub-tasks)

 

 Comments 

 

 

Comment by Dave Ings [ 03/Jun/09 01:06 PM ]

As per 6/3 subteam meeting, assigned to Dieter who will be assisted by Hannah.

Comment by Luc Clement [ 01/Jul/09 10:34 AM ]

Closed with no action




[BP-86] State transitions: Loop on "inProgress" status?

Created: 20/May/09  Updated: 01/Jul/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Oliver Kieselbach (SAP)

Resolution:

Won't Fix

Votes:

0

 

 

 Description 

 

 

Submitted by: Hannah Petereit
Proposed by: sub-task / routing pattern working session team

Target: WS-HT

Issue Description:

WS-HT specification, chapter 4.7 Human Task Behavior and State Transitions
To enable sequential processing of a single task by multiple processors, the status & action model could be enriched by a loop on the "inProgress" status.

Proposal:

Do not add such a loop, as a next processor of a task, which is sequentially processed by multiple people, would usually expect that the task reaches his inbox in status "Reserved". In this case the same state transition applies as for delegation of an "inProgress" task.

 

 Comments 

 

 

Comment by Dave Ings [ 03/Jun/09 01:05 PM ]

As per 6/3 subteam meeting, assigning to Oliver but no work required until the resolution of BP-85.

Comment by Luc Clement [ 01/Jul/09 10:31 AM ]

Closed as obsolete




[BP-85] Sequential routing pattern & sub-tasks

Created: 20/May/09  Updated: 29/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

depends on

BP-92

HT section 3 need new text defining t...

Applied

is depended on by

BP-107

HT 6.1.1 clarify complete operation w...

Closed

 

 Description 

 

 

Submitted by: Hannah Petereit
Proposed by: sub-task / routing pattern working session team

Target: WS-HT

Issue Description:

Does the sequential routing pattern have to be realized via sub-tasks? As an alternative option a single task could be processed by multiple people in a row, by programatically triggering delegates on the task. The proposal would be to enable both implementation options.

Proposal from June FTF (approved):

Dieter Koenig: Mike R. moves that sequential routing patterns must use a separate subtask for each step in a sequential pattern
Dieter Koenig: Dieter seconds
Phil Allen: Motion to amend
Phil Allen: Replace the second sentence of the following with text as proposed by Michael (section 4.6.1.2,
paragraph before 'syntax example'

   After each user completes the task, the next set of users/groups in htd:sequence SHOULD see the
   task until the sequence pattern completes. The same task instance SHOULD be assigned to each
   of the users/groups in sequence.

  Sequential routing patterns MUST use a separate subtask for each step in a sequential pattern.

 

 Comments 

 

 

Comment by Dave Ings [ 03/Jun/09 12:54 PM ]

As per 6/3 subteam meeting Ravi will partner with Gerhard on this issue.

Comment by Dave Ings [ 03/Jun/09 01:08 PM ]

For the task routing patterns work, this is a "keystone" issue and its resolution is fundamental to resolving the other task routing patterns issues.




[BP-84] Completion behavior

Created: 20/May/09  Updated: 29/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Hannah Petereit
Proposed by: sub-task / routing pattern working session team

Target: WS-HT

Issue Description:

Introduce a completion behavior concept & syntax that allows to define completion criteria for a task as well as effects / results when the criteria is met.

Additional remark: "Completion behavior" could be an independent concept that may not only be used by routing patterns / sub-tasks, but gets introduced for all kind of tasks -> enable scenarios where (manual) completion of simple tasks become enabled when a given criterion is met

Proposal (by Dieter König):
http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200905/msg00025.html

 

 Comments 

 

 

Comment by Dave Ings [ 03/Jun/09 01:02 PM ]

As per the 6/3 subteam meeting, Dieter asked that all TC members but in particular members of the TRP subteam review his latest proposal for BP-84. Depending on agenda availability we'll discuss the proposal either at the 6/10 TC or at the 6/17 F2F.




[BP-83] Override capabilities for a B4P peopleActivity calling a task defined with a routing pattern or composite content

Created: 20/May/09  Updated: 02/Sep/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Hannah Petereit
Proposed by: sub-task / routing pattern working session team

Target: B4P & WS-HT

Issue Description:

WS-HT specification, chapter 4.5 Elements for People Assignment
Proposed people assignment syntax (see BP-28)

<htd:potentialOwner>
  fromPattern+
</htd:potentialOwner>

The potentialOwners element must be overwritable on people activity level. Putting fromPattern expressions (-> indicating a routing pattern task) inside the potentialOwner element will make the fact, whether a task is a routing pattern task or not, overwritable as well.

The following scenarios should be supported:
a) Human task definition defines a routing pattern task, but potential owner assignment should be context-based
b) Human task definition defines a simple task, may be "transformed" into a routing pattern task via overwrite mechanisms; -> then also output aggregation needs to be defined via overwrite mechanisms

Proposal: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200908/msg00022.html

 

 Comments 

 

 

Comment by Dave Ings [ 03/Jun/09 12:58 PM ]

As per the 6/3 subteam meeting, Alex will partner with Dieter and Ravi on this issue.




[BP-82] Actual Owner cardinality

Created: 20/May/09  Updated: 28/Oct/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Hannah Petereit
Proposed by: sub-task / routing pattern working session team

Target: WS-HT

Issue Description:

WS-HT specification, chapter 3.1 Generic Human Roles
"A task managed by a WS-HumanTask Processor MUST have exactly one actual owner."

-> Discuss validity concerning routing patterns

 

 Comments 

 

 

Comment by Dave Ings [ 03/Jun/09 12:57 PM ]

As per 6/3 subteam meeting, we agreed that this issue resolution will fall out of the resolutions of the other task routing pattern issues, so there is no need to assign it to anyone yet.

Comment by Dave Ings [ 10/Jun/09 02:47 PM ]

Assigned to Martin as per 6/10 TC discussion.




[BP-81] Task Activation Deferral missing in Human Task Context

Created: 14/May/09  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter König

Target:
ws-humantask-context.xsd

Reference: http://lists.oasis-open.org/archives/bpel4people/200905/msg00019.html

Description:
According to WS-HT chapter 7, "The amount of time by which the task activation is deferred" can be overridden. However, this capability is missing in the WS-HT request context (note that the BP-63 resolution split the human task context into request context and response context).

Proposal:
Add a new element "activationDeferralTime" to the WS-HT request context (see attached version of ws-humantask-context.xsd).

 

 Comments 

 

 

Comment by Luc Clement [ 27/May/09 11:01 AM ]

Accepted as resolved on 27 May 2009

Comment by Ivana Trickovic [ 03/Jun/09 05:34 AM ]

The issue resolution applied in CD-03, Rev6.




[BP-80] Specify default behavior of simple query operations

Created: 15/Apr/09  Updated: 10/Jun/09  Due: 01/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Hannah Petereit

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Hannah Petereit

TARGET: WS-HT specification, chapter 6.1.2 Simple Query Operations

DESCRIPTION:

For both simple query operations (getMyTaskAbstracts & getMyTasks) it is not clearly defined whether input parameters are mandatory, and - in case an input parameter is not set - what is the default behavior of the operation.

PROPOSAL:
1) task type
   optional parameter
   default behavior (in case parameter is not set): "ALL"

2) generic human role
   parameter should be list-enabled
   optional parameter
   default behavior (in case parameter in not set): filter by "processor", which means:
a) "actual owner" for tasks in status "Reserved" / "InProgress" / "Suspended" / "Completed" / "Canceled"
b) "potential owner" for tasks in status "Ready"

3) work queue
   optional parameter
   default behavior (in case parameter is not set): personal tasks only (this is already defined in the spec)

4) status list
   optional parameter
   default behavior (in case parameter is not set): no filtering by status, tasks in all potential states are returned

5) where clause
   optional parameter
   default behavior (in case parameter is not set): no additional task filtering

6) created-on clause
   optional parameter
   default bahavior (in case parameter is not set): no filtering by creation date

7) maxTasks
   optional parameter (this is already defined in the spec)
   default behavior (in case parameter is not set): all result tasks of the query are returned

 

 Comments 

 

 

Comment by Luc Clement [ 13/May/09 12:13 PM ]

At the 13 May 2009 telecon, the following updates where discussed and subsequently BP-80 was approved with these: parameter 2) generic human role should be list-enabled; default behavior (in case no value is passed) will be "actual owner"

Comment by Luc Clement [ 13/May/09 09:30 PM ]

Approved as Resolved 13 May 2009

Comment by Ivana Trickovic [ 03/Jun/09 05:33 AM ]

The issue resolution applied in CD-03, Rev3.




[BP-79] Interoperability between task parents and task processors from different vendors

Created: 04/Mar/09  Updated: 24/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Alex Malek

TARGET: WS-HumanTask

DESCRIPTION:
Following up on the discussion from the F2F, we would like to open an issue tracking the proposal to add support for interoperable "simple tasks" to WS-HT, i.e. enabling task parents to register new simple tasks on a task processor remotely. This would allow task parents and task processors from different vendors to easily interoperate in task scenarios that involve simple tasks.

http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200901/msg00011.html has the original proposal. The attached draft we presented earlier in February during the alternating wed conf calls. We are currently working out another iteration in collaboration with SAP.

See attachment at: http://lists.oasis-open.org/archives/bpel4people/200903/ppt00000.ppt, attached to message http://lists.oasis-open.org/archives/bpel4people/200903/msg00008.html

PROPOSAL:

 

 Comments 

 

 

Comment by Dave Ings [ 04/Mar/09 03:05 PM ]

Assigned to Alex as per 3/4 TC minutes.

Comment by Dave Ings [ 03/Jun/09 12:55 PM ]

As per 6/3 subteam meeting Alex will partner with Oliver to move this issue forward.

Comment by Luc Clement [ 24/Jun/09 10:24 AM ]

Final proposal approved 20090624 as specified at http://lists.oasis-open.org/archives/bpel4people/200906/msg00074.html




[BP-78] Application of conformance targets in WS-HT may require

Created: 02/Mar/09  Updated: 30/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Phil Allen

There are some locations in the WS-HT spec that may require additional changes for applying the conformance targets
RE: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200902/msg00003.html

Please see http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200902/doc00001.doc, which contains the details of the possible locations for updates in the WS-HT spec. There are two of these; there are additional notes for RFC2119, which are handled separately.

 

 Comments 

 

 

Comment by Dave Ings [ 04/Mar/09 03:04 PM ]

Luc to reassign as per 3/4 TC minutes.




[BP-77] Application of RFC2119 to WS-HT may require some updates

Created: 02/Mar/09  Updated: 30/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Phil Allen

There are some locations in the WS-HT spec that may require additional changes for RFC 2119.
RE: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200902/msg00003.html

Please see http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200902/doc00001.doc, which contains the details of the possible locations for updates in the WS-HT spec. It contains (also) two elements for conformance target updates (handled separately)




[BP-76] Application of RFC2119 rules in B4P may require some updates

Created: 02/Mar/09  Updated: 30/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ralf Mueller

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Phil Allen

There are some locations in the B4P spec that may require additional changes for RFC 2119.
RE: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200902/msg00003.html

Please see http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200902/doc00000.doc, which contains the details of the possible locations for updates in the B4P spec.




[BP-75] Clarification and/or Definition of Some Technical Terms and Statements

Created: 23/Feb/09  Updated: 16/May/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Alireza Farhoush

TARGET: WS-HumanTask Specification Version 1.1 CD02

DESCRIPTION:
In Section 3.2 'Assigning People', in the 2nd paragraph, a statement reads:

"Using htd:tOrganizationalEntity allows to assign either a set of people or an unresolved group of people ('work queue')."

The term "Unresolved group of people ('work queue')" should be defined.
Also, the statement above does not seem to read correctly. The verb "allows" makes one expect a noun (grammatically speaking, a direct object) to follow.

PROPOSAL (L Clement):
The first two paragraphs of Section 3.2 should read as:

3.2 Assigning People
To determine who is responsible for acting on a human task in a certain generic human role or who will receive a notification, people need to be assigned. People assignment can be achieved in different ways:
• Via logical people groups (see 3.2.1 "Using Logical People Groups")
• Via literals (see 3.2.2 "Using Literals")
• Via expressions e.g., by retrieving data from the input message of the human task (see

3.2.3 "Using Expressions").
When specifying people assignments then the data type htt:tOrganizationalEntity is used. <replacementText>The htt:tOrganizationalEntity element specifies the people assignments associated with generic human roles used. </replacementText>

---------------------


DESCRIPTION:
In Section 3.2.1 'Using Logical People Groups', in the 1st paragraph, a statement reads:

"During people query execution an infrastructure may decide which of the parameters defined by the logical people group are used."

The term "infrastructure" is too vague and general and should be concretely defined.

DISPOSITION (L Clement):
This text has been handled by BP-77 which replaced "infrastructure" by "task processor". CD-03 will contain the following change to section 3.2.1 which will state:

3.2.1 Using Logical People Groups
A logical people group represents one person, a set of people, or one or many unresolved groups of people (i.e., group names). A logical people group is bound to a people query against a people directory at deployment time. Though the term query is used, the exact discovery and invocation mechanism of this query is not defined by this specification. There are no limitations as to how the logical people group is evaluated. At runtime, this people query is evaluated to retrieve the actual people assigned to the task or notification. Logical people groups MUST support query parameters which are passed to the people query at runtime. Parameters MAY refer to task instance data (see section 3.4 for more details). During people query execution a task processor can decide which of the parameters defined by the logical people group are used. A WS-HumanTask Processor MAY use zero or more of the parameters specified. It MAY also override certain parameters with values defined during logical people group deployment. The deployment mechanism for tasks and logical people groups is out of scope for this specification.

---------------------------------------


DESCRIPTION:
In Section 3.2.1 'Using Logical People Groups', in the 5th paragraph, a statement reads:

"Especially in cases where group membership changes frequently, this 'late binding' to the actual group members is beneficial."

 "This 'late binding'" is an ambiguous antecedent and needs to specify that late binding is a reference to "the creation of a human task or a notification"

PROPOSAL (L Clement):

Change:
People queries return either one person, a set of people, or the name of one or many groups of people. The latter is added to support "work queue" based business scenarios, where people see work they have been assigned to due to their membership of a certain group. Especially in cases where group membership changes frequently, this "late binding" to the actual group members is beneficial.

To:
People queries return one person, a set of people, or the names of one or more groups of people. The use of a group enables the ability to create a human "work queue" where members are provided access to work items assigned to them as a result of their membership of a group. The ability to defer group membership is beneficial when group membership changes frequently.

----------------------------------------------

DESCRIPTION:
In Section 3.3 'Task Rendering', in the 1st paragraph, a statement reads:

"The key elements are the task list client, the task engine and the applications invoked when a task is executed."

The term "task engine" is not defined elsewhere in the document. Should it be replaced with task processor, which has been previously defined?

PROPOSAL (L Clement):
Change "task engine" to "task processor".

The first paragraph of Section 3.3 Task Rendering will then state:

Humans require a presentation interface to interact with a machine. This specification covers the service interfaces that enable this to be accomplished, and enables this in different constellations of software from different parties. The key elements are the task list client, the task processor and the applications invoked when a task is executed.

 

 Comments 

 

 

Comment by Dave Ings [ 04/Mar/09 03:10 PM ]

During the 3/4 TC we agreed it would feel very odd to directly refer this to the editor's SC. So Luc will draft the revised spec text and bring it back to the TC as a formal proposal.

Comment by Luc Clement [ 29/Apr/09 10:34 AM ]

changed applied by Luc Clement




[BP-74] Single Actual Owner and Routing Tasks

Created: 23/Feb/09  Updated: 04/Mar/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

Duplicate

Votes:

0

 

 

 Description 

 

 

Submitted by: Alireza Farhoush

TARGET: WS-HumanTask Specification Version 1.1 CD02

DESCRIPTION:
In Section 3.1 'Generic Human Roles', in the 6th paragraph, a statement reads:

"A task managed by a WS-HumanTask Processor must have exactly one actual owner."

Although the routing pattern has not been finalized, it does raise an issue regarding the above statement. In routing patterns, conceptually, multiple people might be invited to work on the same task, although, in most implementations, a new instance of the task is created and there is therefore only one owner per instance. Perhaps some clarification about single ownership of a task should be added to accommodate the semantics of routing tasks (once approved).

 

 Comments 

 

 

Comment by Luc Clement [ 04/Mar/09 10:54 AM ]

We decided during the 4 Mar call to incorporate this as part of issue BP-28. We agreed to close the issue.

Comment by Luc Clement [ 04/Mar/09 10:55 AM ]

Input to BP-28




[BP-73] Priority of a Task and Potential Owner Privilege

Created: 23/Feb/09  Updated: 29/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Alireza Farhoush

TARGET: WS-HumanTask Specification Version 1.1 CD02

DESCRIPTION:
In Section 3.1, 'Generic Human Roles', in the 4th paragraph, a statement reads:

"...potential owners can influence the progress of the task, for example by changing the priority of the task."

Shouldn't changing a task priority be performed only by the stakeholder or the business administrator? Multiple potential owners could independently modify the priority as they see fit (and perhaps never acquire a task).

PROPOSAL:
Limit the privilege of changing task priority to Task Stakeholder and Business Administrator.

Proposal posted at http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200906/msg00017.html




[BP-72] Add the definition of the Human Task Process term

Created: 18/Feb/09  Updated: 16/May/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Yoichi Takayama

Target: bpel4people-1.1-spec-CD-02-MARKEDUP.pdf

Description:

This document mentions HumanTask Processor only 3 times in the Section 4.5.3.

920 A WS-HumanTask Processor MUST ...

928 HumanTask Processor hosting ...

948 ... WS-HumanTask
949 Processor ...

While the definition for a BPEL Processor is given in Section 2.3 as below, there appears to be no clear definition or reference to what the HumanTask Processor is at this point (although the reader would find it out in WS-HumanTask document later).

251 BPEL4People Processor
252 A BPEL4People Processor is any implementation that accepts a BPEL4People definition and
253 executes the semantics defined in this document.

Except that The HumanTask Processor must also return a metadata, and the metadata operation is said to refer Section 8.2 of the WS-HumanTask document.

It seems that, before introducing a new term, it is better to define it. Although the usage of the term is very limited in this case, the term is neither common nor self-explanatory to someone who reads this document (or works with this specification) for the first time.

Proposal:

Add the definition of the term at the beginning of the block or immediately after the first usage. Maybe refer Section 8.2 of the WS-HumanTask document at the very first time it was used.

Also, in addition, clarify what it is briefly in this document. It would be helpful because then the reader would understand what it means without referring to the WS-HumanTask document unless more clarification is needed.

Notes:

In my opinion it is better to define a Task List Manager sub-system (or Task Manager) who receives a Task request from BPEL4People Process definition processing engine. Task Processor sounds like it "processes" the Task, which is to interpret the XML definition and create the instance of a job entry in the Task Management sub-system. Then, the Task Management sub-system must handle all Task entries afterwords for creating notices to the end users (if the Tasks were already assigned) or put them on the "Claimable Jobs" Task List, handle Potential Owners claiming the job and becoming the Actual Owner, or do whatever the life-cycle management.

At this stage, none of the Web Services is involved yet.

When an Actual Owner executes a Task (actually only using the Task job entry at this point), the Task Manager contacts the Web Service specified in the "interface" of the Task, gets the Task UI from the Web Service and makes a Task Client (which is at its disposal) to present the Task UI to the end user.

 

 Comments 

 

 

Comment by Luc Clement [ 16/Apr/09 07:05 AM ]

Proposed disposition (by Luc Clement)

Update section 2.3 Conformance Targets by adding the following two bullets:

• WS-HumanTask Definition
A WS-HumanTask Definition is any artifact that complies with the human interaction schema and additional constraints as defined by the WS-HumanTask 1.1 specification.
• WS-HumanTask Processor
A WS-HumanTask Processor is any implementation that accepts a WS-HumanTask definition and executes the semantics as defined by the WS-HumanTask 1.1 specification.

Comment by Luc Clement [ 29/Apr/09 10:19 AM ]

Approved 29 Apr by the TC as resolved based on the proposal created by Luc Clement

Comment by Luc Clement [ 29/Apr/09 10:35 AM ]

changed applied by Luc Clement




[BP-71]  SearchBy and PrimarySearchBy

Created: 05/Feb/09  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter König

Target: WS-HumanTask 1.1 CD 02, ws-humantask-types.xsd

Description: WS-HT 1.1 section 4.2 introduces the task property "searchBy", "... used to search for task instances based on a custom search criterion".

Section 3.4.4 defines a data type for task instance data, and sections
6.1.2 / 6.1.3 define query result set data types, both containing a field "primarySearchBy".

There appears to be no reason to have different names.

Proposal: In WS-HumanTask 1.1 CD 02 and ws-humantask-types.xsd, rename "primarySearchBy" to "searchBy".




[BP-70] Complement simple query operations

Created: 21/Jan/09  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ralf Mueller

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Relates to

relates to

BP-45

Task History

Applied

 

 Description 

 

 

SUBMITTED BY: Gerhard Pfau

TARGET: WS-HumanTask

DESCRIPTION:
When comparing the simple query operations in "6.1.2 Simple Query Operations" with the advanced query operation in "6.1.3 Advanced Query Operation" we realized that the simple queries are lacking of two basic capabilities:
a) Support for paging, i.e., get subsets of a query result set of a certain length that starts either at an offset of 0 or with an arbitrary offset
b) Support for ordering the result

PROPOSAL:
Offer the following parameters from the advanced query operation also for the simple queries with semantics as described in "6.1.3 Advanced Query Operation":
• order-by clause
• taskIndexOffset

     Gerhard Pfau




[BP-69] Supporting Batch Operations

Created: 21/Jan/09  Updated: 29/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Relates to

relates to

BP-44

Discovering the operations available ...

Applied

 

 Description 

 

 

SUBMITTED BY: Gerhard Pfau

TARGET: WS-HumanTask

DESCRIPTION: Task list clients often deal with many human tasks, performing the same action on them at the same time, e.g., when a user operates on a selection of several tasks in a UI application. Today in WS-HT this is accomplished by invoking a particular operation for each single human task of such list. When dealing with larger numbers of human tasks this involves a lot of communication between client and server and thus should be optimized. For example instead of a claim operation that takes a single task id as an input we may want to offer a batch claim operation that processes a set of task ids in a single operation.

PROPOSAL / THOUGHTS: Analyse the operations in chapter "6.1 Operations for Client Applications", and particularly "6.1.1 Participant Operations", and define which operations are likely to be called for a set of human tasks and deserve optimization by offering a corresponding batch operation. Specify these batch operations (input / output / fault messages). Agree on a naming scheme for batch operations that makes clear which original operation the are derived from. .

WSDL posted 20090610 with post: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200906/msg00015.html




[BP-68] htd:remoteTask - a possible error?

Created: 20/Jan/09  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Raised By: Yoichi Takayama

Target: WS-HumanTask, V 1.1, CD02, Jan 6 2009

http://www.oasis-open.org/apps/org/workgroup/bpel4people/document.php?
document_id=30552

Description:

Line 2397 states htd:remoteTask/@responseOperation attribute. As far as I know, the remoteTask is a b4p element. Is this an error?

Proposal:

Change htd to b4p.

 

 Comments 

 

 

Comment by Dave Ings [ 30/Jan/09 03:40 PM ]

Opened as per 2009-01 F2F minutes.

Comment by Dave Ings [ 30/Jan/09 03:41 PM ]

As per 2009-01 F2F minutes, agreed to resolution was:

Resolve 68 by deleting the sentence at 2298 "The value of this element is taken from the htd:remoteTask/ @responseOperation attribute." in HT.




[BP-67] Clarify the words, Requesting Application, Task Parent, the Application Logic and the Task in Section 7.

Created: 20/Jan/09  Updated: 30/Jan/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

Duplicate

Votes:

0

 

 

 Description 

 

 

Raised By: Yoichi Takayama

Target: V 1.1, CD02, Jan 6 2009

http://www.oasis-open.org/apps/org/workgroup/bpel4people/document.php?
document_id=30552

Description:

The "Task" talked about in this section seems to indicate that it is a Task Implementation, i.e. a Task application or a Web Service that implements it, not a Task representation on the BPEL4People application. The latter is subject to the life cycle management and e.g. involved on who claims it etc., but the former, the Task application as a Web Service implementation, does not need to know anything but who uses the Web Service to run a Task application.

The former is associated with the Task List Client, and the latter, with Task Client.

Assuming that the Coordinator indicated in Figure 1 is the WS-C Coordinator, the meaning of the Requesting Application, Task Parent, the Application Logic and the Task must be all clarified.

Proposal:

Task is the Web Service that may be invoked as a Human Task implementation. This is defined in <htd:task> or <b4p:remoteTask>. When the TaskProcessor requires the Task application instance, it invokes it to create the Task application instance. When the TaskProcessor needs to pass the UI of the Task application to a corresponding Task Client, it uses WS-C to send a "StartTask" message.

Coordinator: This is a WS-C Coordinator that sits between the Task Parent and the Task. The Task is implemented as WS-C-enabled Web Service. The Coordinator is the registration service of WS-C Protocol Service EPRs of two parties which wish to communicate during a single Web Service invocation (i.e. a session). Each Task instance creates a separate EPR exchange with a unique instance ID.

Requesting Application: In this case, it is a Task Life Cycle management application. In case of CD02, this seems to be designated as Task Processor. However, from Task point of view, this is a combination of BPEL engine and Task List application (the Task Processor does all the leg work for the Task List application).

Task Parent: This word is ambiguous. Each Task instance (application implementation on the Service side) may have a corresponding Task Client to work with on the BPEL4System. This may be called a Task Parent. It is a separate concept from the Requesting Application. Maybe Task Parent should be included inside the Application Logic (or behind it). Probably Task List Application is more appropriate, although I prefer Task Engine or Task Manager.

Application Logic: In this case, the Application logic is the hard-wired logic of the Task Processor, dictating what WS-C messages are to be sent to the Task. In a larger picture, however, this is all the hard/soft logics of Process, Process engine, Task List application and the Task Processor. In a narrow sense, however, Application Logic regarding the communication can be actually dissected to 3 parts. Web Service communication, WS-C communication and Task application communication. In this diagram, the Application Logic is the combination of all these and may be the diagram should be refined with different components involved. For example, the Coordinator does not really handle the messages (1) and (4a or 4b).

What is left out of Figure 1 is another application layer. That is, the Task Client, that receives the Task application from the Web Service and runs it for the end user, if it is a, inline or a local Task.

For a remoteTask, there is no human interface needed and no Task Client gets involved. All human users are expected on the other side of the Service. This is a black box in terms of the human operation. For a remoteTask, the WS-C still must operate to deal with the asynchronous nature of the human task request.

 

 Comments 

 

 

Comment by Dave Ings [ 30/Jan/09 03:39 PM ]

Close as duplicate of BP-65 as per 2009-01 F2F minutes.




[BP-66] Delete or modify the mention/usage of the TaskProcessor in Section 7 and Appendix A of WS-HumanTask

Created: 20/Jan/09  Updated: 30/Jan/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

 

 Description 

 

 

Raised By: Yoichi Takayama

Target: V 1.1, CD02, Jan 6 2009

http://www.oasis-open.org/apps/org/workgroup/bpel4people/document.php?
document_id=30552

Description:

The TaskProcessor does not receive the messages over WS-C. The mention and the usage of TaskProcessor in conjunction with Task implemented as Web Service are completely wrong.

Proposal:

Move the TaskProcessor to the Application side of the diagram. The TaskProcessor must communicate with the WS-HT Protocol Service on the Web Service that implemented the Task. Also, the (WS-C) WS-HT Protocol Service is the proper term used in WS-C, that handles the Register events to exchange the EPRs for the WS-C interface.

 

 Comments 

 

 

Comment by Dave Ings [ 30/Jan/09 03:37 PM ]

Closed without action as per 2009-01 F2F minutes.




[BP-65] Add a clear proposed system component architecture dictated by the BPEL4People and WS-HumanTask

Created: 20/Jan/09  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

File Attachments:

Description: FileB4P Arch.pptx    

 

 Description 

 

 

Raised By: Yoichi Takayama

Target: V 1.1, CD02, Jan 6 2009

http://www.oasis-open.org/apps/org/workgroup/bpel4people/document.php?
document_id=30552

Description:

The Task life cycle management API and the Task Web Service interface must be clearly distinguished as V 1.0. In CD02, however, there seems to be a confusion about the demarcation and the role of each element.

Proposal:

Add a prospective system component architecture dictated by this standard and attach the relevant interface used to each part.

 

 Comments 

 

 

Comment by Dieter Koenig [ 21/Jan/09 06:53 AM ]

Overview picture to be used for the issue resolution

Comment by Dave Ings [ 30/Jan/09 03:36 PM ]

Opened and assigned to Frank as per the 2009-01 F2F minutes.

Comment by Luc Clement [ 27/May/09 10:47 AM ]

We revolve BP-65 during the 27 May 2009 telecon based on the latest draft offered by Frank L: http://www.oasis-open.org/apps/org/workgroup/bpel4people/download.php/32646/HT%20Architecture%20v0-9-2.docx




[BP-64] A need for a "GetUI" or "StartTask" request/message for a Task

Created: 19/Jan/09  Updated: 15/Apr/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Yoichi Takayama

Resolution:

Fixed

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Yoichi Takayama

Target: B4P / HT

WS-BPEL Extension for People (BPEL4People) Specification V 1.1, CD 02
(proposal), 6 Jan 2009, Section 6 (and related sections)

Also:

Web-Services Human Task (WS-HumanTask) Specification Version 1.1, CD
02 (proposal), 6 Jan 2009, Section 7 (and related sections)

http://www.oasis-open.org/apps/org/workgroup/bpel4people/document.php?
document_id=30552


Description:

In WS-HumanTask Section 7, Figure 1, the requestMessage presumably
corresponds to createTask message of the Task Life Cycle management.
Although this causes to "Register" the WS-C protocol handler EPRs
between the Task Parent and the Task (implemented as a Web Service)
as the (2) and (3) indicate, the "requestMessage is an asynchronous
request and nothing follows after that. The (4a) is the asynchronous
return for the (1) requestMessage, as I understand it.

Presumably, following this step, there is the "start" message handled
by the Task Life Cycle management (BPEL4People 4.8 or WS-HumanTask
4.7), which presumably would get UI from the Task Implementation to
the Task Parent (and then to a Task Client and to its user agent that
the end human user uses to do the Task) and let the end user do the
human task. However, how this would be dealt with in an interoperable
manner is not shown by the specification.

Obviously, the (4a) or (4b) would not occur until the Task UI is
presented to the end user and some outcome has been produced. So, the
"start" step is a Task Service step that should be documented and
defined before the termination of the Task such as (4a) or (4b). This
is lacking in the specification.

We cannot presume that the Task Parent or the Task Coordinator knows
what operation or EPR to ask for to call it, since the specification
does not have any place that states such an operation or how to
specify such additional EPR for that operation, i.e. "StartTask".

Proposal:

1. Add a "StartTask" request message to Section 6.1 of the WS-
HumanTask and to the table of WS-C WS-HT messages (page 31). This
corresponds to the "start" Task Life Cycle management command, but is
distinct from it because it is one of the Task Web Service massages.

This may return either a single page UI (such as HTML Form or XML) or
an application UI (that could be a form of an EPR or URL or e.g. XUL-
application XML launch point). The Task may be an multiple-page Web
application or Desktop UI of an comprehensive application, for example.

2. For this reason, the interoperable format for the response message
of the "StartTask" must be defined so that it can be interpreted by
the Task Coordinator, Task Parent, and Task Client properly (so that
they will know what the response contains and what to do with the
information/data).

Alternative

We could provide an ability or a place to state the additional EPR
for StartTask, but that is tedious. If a fixed WS-C WS-HT message can
handle it, it is more consistent with the current CD and it may be
simpler.

Potential Problem

If a Web Service implements multiple Tasks with different EPRs, each
Task must implement different WS-C protocol handler and the
"StartTask" message handler with it. In this case, there may be more
efficient way to organize the whole WS-C WS-HT better.

 

 Comments 

 

 

Comment by Luc Clement [ 15/Apr/09 11:06 AM ]

This was closed with no action on 15 April 2009

The last message on this topic was: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200904/msg00017.html




[BP-63] Actual Owner Missing in WS-HumanTask Context

Created: 17/Dec/08  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter König

Target: WS-HumanTask 1.1 specification, section 7.4 Providing Human Task Context, ws-humantask-context.xsd

Description: BPEL4People provides the getActualOwner XPath extension function. Other WS-HumanTask Parent applications may be interested in the actual owner of a task as well. However, the actual owner of a task is currently not communicated back to the calling WS-HumanTask Parent.

** NOTE on the proposal - Typographical errors in the original submission where identified during the 18 Feb 09 telecon. Dieter resubmitted the content as attachments. See attachment to: http://lists.oasis-open.org/archives/bpel4people/200902/msg00015.html **

Proposal: Add a child element "actualOwner" of type "htt:tUser" to the WS-HumanTask context data type "tHumanTaskContext".

   <xsd:complexType name="tHumanTaskContext">
      <xsd:sequence>
         <xsd:element name="priority" type="htt:tPriority" minOccurs="0"/>
         <xsd:element name="peopleAssignments" type="tPeopleAssignments" minOccurs="0"/>
         <xsd:element name="isSkipable" type="xsd:boolean" minOccurs="0"/>
         <xsd:element name="expirationTime" type="xsd:dateTime" minOccurs="0"/>
         <xsd:element name="outcome" type="xsd:string" minOccurs="0"/>
         <xsd:element name="attachments" type="tAttachments" minOccurs="0"/>
         <xsd:element name="actualOwner" type="htt:tUser" minOccurs="0"/>
      </xsd:sequence>
   </xsd:complexType>

Alternative Proposal: Split the WS-HumanTask context into an inbound and an outbound context, and add the actual owner to thr outbound context.

   <xsd:complexType name="tHumanTaskRequestContext">
      <xsd:sequence>
         <xsd:element name="priority" type="htt:tPriority" minOccurs="0"/>
         <xsd:element name="peopleAssignments" type="tPeopleAssignments" minOccurs="0"/>
         <xsd:element name="isSkipable" type="xsd:boolean" minOccurs="0"/>
         <xsd:element name="expirationTime" type="xsd:dateTime" minOccurs="0"/>
         <xsd:element name="attachments" type="tAttachments" minOccurs="0"/>
      </xsd:sequence>
   </xsd:complexType>

   <xsd:complexType name="tHumanTaskResponseContext">
      <xsd:sequence>
         <xsd:element name="outcome" type="xsd:string" minOccurs="0"/>
         <xsd:element name="attachments" type="tAttachments" minOccurs="0"/>
         <xsd:element name="actualOwner" type="htt:tUser" minOccurs="0"/>
      </xsd:sequence>
   </xsd:complexType>

 

 Comments 

 

 

Comment by Dave Ings [ 30/Jan/09 03:08 PM ]

Opened as per 2009-01 F2F minutes.

Comment by Luc Clement [ 18/Feb/09 11:49 AM ]

Typographical errors in the email and the above text where identified during the 18 Feb 09 telecon. Dieter resubmitted the content as attachments. See attachment to: http://lists.oasis-open.org/archives/bpel4people/200902/msg00015.html

Comment by Ivana Trickovic [ 03/Jun/09 05:34 AM ]

The issue resolution applied in CD-03, Rev6.




[BP-62] Blocked Substitution Elements in Human Task Definitions

Created: 13/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask.xsd

Description: The WS-HumanTask language XML Schema contains an abstract element "genericHumanRole" and substitution elements (such as "potentialOwner" etc.) which are used in people assignments in place of the abstract element. However, in a concrete task definition (an XML Schema instance document), the schema currently does not allow this replacement. The reason is the blockDefault="#all" attribute on the schema root element.

Proposal: In general, it is reasonable to block substitutions and derivations by extension/restriction for WS-HT language elements. Both languages are already extensible by allowing elements and attributes of other namespaces to be added to all existing language elements. It is therefore not necessary to allow introducing substitution elements or derived types. The same approach is used in bpel4people.xsd (and in WS-BPEL 2.0 XML Schema artifacts).

In this particular case (genericHumanRole), it is necessary to allow substitutions in instance documents, otherwise people assignments cannot be specified at all. This can be done by overriding the global "blockDefault" setting by using a concrete "block" attribute. This attribute must allow substitutions, but type derivations by extension/restriction may still remain blocked.

As a result, the definition of the "genericHumanRole" abstract element should be changed from:
<xsd:element name="genericHumanRole" type="tGenericHumanRole" abstract="true" />
to:
<xsd:element name="genericHumanRole" type="tGenericHumanRole" abstract="true" block="extension restriction" />



 

 Comments 

 

 

Comment by Dave Ings [ 30/Jan/09 03:05 PM ]

Opened and resolution accepted as per 2009-01 F2F minutes.

Comment by Dieter Koenig [ 20/Feb/09 10:53 AM ]

Issue resolution applied:
   ws-humantask.xsd
      - added block="extension restriction" to abstract genericHumanRole element definition (line 119)
   bpel4people.xsd (!)
      - added block="extension restriction" to abstract genericHumanRole element definition (line 77)




[BP-61] Cleanup tTask Definitions

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: WS-HumanTask 1.1, ws-humantask-types.xsd

Description: Two different 'kinds' of the tTask type exist in WS-HT:
1. In 'ws-humantask.xsd', tTask is part of the WS HT language definition.
2. In 'ws-humantask-types.xsd', the tTask defines the information structure of the result that is delivered when getTaskDetails() or getMyTasks() are invoked.
While this is not an XML Schema or WSDL bug, it is not consistent and therefore confusing.

Proposal: Consistently use the following naming conventions:
- Continue to use tTask as part of the WS-HT language XML Schema (that is, no change in ws-humantask.xsd)
- For the WS-HT data types XML Schema and the WS-HT API, use:
o tTaskAbstract and getMyTaskAbstracts()
o tTaskDetails and getMyTaskDetails(), getTaskDetails()

A: Rename the tTask type and the task element in ws-humantask-types.xsd:
<xsd:element name="taskDetails" type="tTaskDetails" />

<xsd:complexType name="tTaskDetails">
  <xsd:sequence>
    <xsd:element name="id" type="xsd:string" />
    <xsd:element name="taskType" type="xsd:string" />
   ...

B: Rename the getMyTasks() operation in ws-humantask-api.wsdl:
<wsdl:operation name="getMyTasks">
   <wsdl:input message="getMyTasks"/>
   <wsdl:output message="getMyTasksResponse"/>
   <wsdl:fault message="illegalStateFault" name="illegalStateFault"/>
   <wsdl:fault message="illegalArgumentFault" name="illegalArgumentFault"/>
</wsdl:operation>

to:
<wsdl:operation name="getMyTaskDetails">
   <wsdl:input message="getMyTaskDetails"/>
   <wsdl:output message="getMyTaskDetailsResponse"/>
   <wsdl:fault message="illegalStateFault" name="illegalStateFault"/>
   <wsdl:fault message="illegalArgumentFault" name="illegalArgumentFault"/>
</wsdl:operation>

and
<wsdl:operation name="getTaskInfo">
  <wsdl:input message="getTaskInfo"/>
  <wsdl:output message="getTaskInfoResponse"/>
  <wsdl:fault message="illegalArgumentFault" name="illegalArgumentFault"/>
</wsdl:operation>

to:
<wsdl:operation name="getTaskDetails">
  <wsdl:input message="getTaskDetails"/>
  <wsdl:output message="getTaskDetailsResponse"/>
  <wsdl:fault message="illegalArgumentFault" name="illegalArgumentFault"/>
</wsdl:operation>

C: Adapt the corresponding WSDL messages in ws-humantask-api.wsdl from:
<wsdl:message name="getMyTasks">
  <wsdl:part element="getMyTasks" name="getMyTasks"/>
</wsdl:message>

to:
<wsdl:message name="getMyTaskDetails">
  <wsdl:part element="getMyTaskDetails" name="getMyTaskDetails"/>
</wsdl:message>

and from:
<wsdl:message name="getMyTasksResponse">
  <wsdl:part element="getMyTasksResponse" name="getMyTasksResponse"/>
</wsdl:message>

to:
<wsdl:message name="getMyTaskDetailsResponse">
  <wsdl:part element="getMyTaskDetailsResponse" name="getMyTaskDetailsResponse"/>
</wsdl:message>

and from:
<wsdl:message name="getTaskInfo">
  <wsdl:part element="getTaskInfo" name="getTaskInfo"/>
</wsdl:message>

to:
<wsdl:message name="getTaskDetails">
  <wsdl:part element="getTaskDetails" name="getTaskDetails"/>
</wsdl:message>

and from:
<wsdl:message name="getTaskInfoResponse">
  <wsdl:part element="getTaskInfoResponse" name="getTaskInfoResponse"/>
</wsdl:message>

to:
<wsdl:message name="getTaskDetailsResponse">
  <wsdl:part element="getTaskDetailsResponse" name="getTaskDetailsResponse"/>
</wsdl:message>

D: Adapt the input/output wrapper elements in ws-humantask-api.wsdl from:
<xsd:element name="getMyTasks">
<xsd:complexType>
          <xsd:sequence>
...

to:
<xsd:element name="getMyTaskDetails">
<xsd:complexType>
          <xsd:sequence>

and from:
<xsd:element name="getMyTasksResponse">
<xsd:complexType>

to:
<xsd:element name="getMyTaskDetailsResponse">
<xsd:complexType>

and from:
<xsd:element name="getTaskInfo">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="identifier" type="xsd:anyURI"/>
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>

to:
<xsd:element name="getTaskDetails">
  <xsd:complexType>
    <xsd:sequence>
        <xsd:element name="identifier" type="xsd:anyURI"/>
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>

and from:
<xsd:element name="getTaskInfoResponse">
   <xsd:complexType>
<xsd:sequence>
           <xsd:element name="task" type="htt:tTask"/>
        </xsd:sequence>
   </xsd:complexType>
</xsd:element>

to:
<xsd:element name="getTaskDetailsResponse">
   <xsd:complexType>
<xsd:sequence>
           <xsd:element name="taskDetails" type="htt:tTaskDetails"/>
        </xsd:sequence>
   </xsd:complexType>
</xsd:element>


 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:53 AM ]

Issue resolution applied:
   ws-humantask-types.xsd
      - renamed the tTask type and the task element to tTaskDetails and taskDetails, respectively (lines 86-87)
   ws-humantask-api.wsdl
      - renamed getMyTasks to getMyTaskDetails (multiple occurences: elements, messages, operation)
      - renamed getTaskInfo to getTaskDetails (multiple occurences: elements, messages, operation)
   ws-humantask-1.1-spec.doc
      - renamed the tTask type and the task element to tTaskDetails and taskDetails, respectively, in 3.4.4, 6.1.1, 6.1.2, 6.1.5
      - renamed getMyTasks to getMyTaskDetails in 6.1.2 and 6.1.5
      - renamed getTaskInfo to getTaskDetails in 6.1.1 and 6.1.5
      - minor formatting (3x lower case operation names) in 6.2




[BP-60] Use tFault in setFault and fail API Methods

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: WS-HumanTask 1.1, ws-humantask-api.wsdl (Follow-up from Issue 46)

Description: In the resolution of Issue 46, the wrapper type tFault (containing a fault name and fault data) was introduced as return value for the getFault() API operation.

The setFault() and fail() operations also use the faultName and faultData elements (as input parameter).

Proposal: Consistently use the tFault wrapper type, that is, replace occurrences of two individual fault-related parameters by one parameter of type tFault.

A: In WS-HumanTask 1.1, section 6.1.1 "Participant Operations", table entry "setFault" and "fail", column "Parameters", paragraph "In", replace the two bullets "faultName" and "faultData" by a single bullet "fault - contains the fault name and the fault data".

B: In "ws-humantask-api.wsdl", replace the "setFault" and "fail" input wrapper element definitions in order to use the newly created tFault type instead faultName and faultData.

Replace:
<xsd:element name="setFault">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="identifier" type="xsd:anyURI"/>
      <xsd:element name="faultName" type="xsd:NCName"/>
      <xsd:element name="faultData" type="xsd:anyType"/>
    </xsd:sequence>
</xsd:complexType>
</xsd:element>

by:
<xsd:element name="setFault">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="identifier" type="xsd:anyURI"/>
      <xsd:element name="fault" type="htt:tFault"/>
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>
  
Replace:
<xsd:element name="fail">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="identifier" type="xsd:anyURI"/>
      <xsd:element minOccurs="0" name="faultName" type="xsd:NCName"/>
      <xsd:element minOccurs="0" name="faultData" type="xsd:anyType"/>
    </xsd:sequence>
</xsd:complexType>
</xsd:element>

by:
<xsd:element name="fail">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="identifier" type="xsd:anyURI"/>
      <xsd:element minOccurs="0" name="fault" type="htt:tFault" />
    </xsd:sequence>
</xsd:complexType>
</xsd:element>


 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:51 AM ]

Issue resolution applied:
   ws-humantask-api.wsdl
      - replaced faultName/faultData by complex typed element fault in setFault and fail operations (lines 186 and 500)
   ws-humantask-1.1-spec.doc
      - replaced faultName/faultData by complex typed element fault in setFault and fail operations in 6.1.1
      - minor formatting (3x lower case operation names) in 6.1.1




[BP-59] MaxOccurs Constraint Wrong in tLiteral

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask.xsd

Description: In WS-HumanTask 1.1, section 3.2.2, the people assignment using a literal requires that exactly one element (of type "tOrganizationalEntity") is specified as the "from"-value. It is not allowed to specify multiple literal elements.

Proposal: In ws-humantask.xsd, drop maxOccurs="unbounded" from the element wildcard in the "tLiteral" definition as shown below (that is, let it default to maxOccurs="1"):

<xsd:complexType name="tLiteral" mixed="true">
  <xsd:sequence>
    <xsd:any namespace="##any" processContents="lax" minOccurs="0" />
  </xsd:sequence>
  <xsd:anyAttribute namespace="##other" processContents="lax" />
</xsd:complexType>


 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:48 AM ]

Issue resolution applied:
   ws-humantask.xsd
      - dropped minOccurs="0" and (!) maxOccurs="unbounded" from tLiteral any particle (line 450)




[BP-58] Type of Attribute LogicalPeopleGroup within tFrom Element

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask.xsd

Description: Within the tFrom complexType definition, the attribute LogicalPeopleGroup must be of type NCName, not QName. The WS-HumanTask specification requires all LPGs to be either defined or redefined within an htd:humanInteractions element - therefore, their names are unique and they can be referenced by NCName. Note that this is already reflected in the spec itself.

Proposal: Within the tFrom complexType definition, change the LogicalPeopleGroup attribute definition from:

<xsd:attribute name="logicalPeopleGroup" type="xsd:QName" />

to:

<xsd:attribute name="logicalPeopleGroup" type="xsd:NCName" />


 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:47 AM ]

Issue resolution applied:
   ws-humantask.xsd
      - changed attribute logicalPeopleGroup definition from QName to NCName (line 428)




[BP-57] Common Base Type for tQuery and tExpression

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask.xsd

Description: The XML Schema type definitions for "tQuery" and "tExpression" both contain element and attribute wildcards for their open content and extensibility. Both can not carry a documentation element.

Proposal: In ws-humantask.xsd, let both types extend the common extension base "tExtensibleMixedContentElements" in order to allow for documentation child elements. As that base type already has the element and attribute wildcards, the definition of the two types can be simplified as shown below:

  <xsd:complexType name="tQuery" mixed="true">
    <xsd:complexContent>
      <xsd:extension base="tExtensibleMixedContentElements">
        <xsd:attribute name="part" />
        <xsd:attribute name="queryLanguage" type="xsd:anyURI" />
      </xsd:extension>
    </xsd:complexContent>
  </xsd:complexType>

  <xsd:complexType name="tExpression" mixed="true">
    <xsd:complexContent>
      <xsd:extension base="tExtensibleMixedContentElements">
        <xsd:attribute name="expressionLanguage" type="xsd:anyURI" />
      </xsd:extension>
    </xsd:complexContent>
  </xsd:complexType>


 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:47 AM ]

Issue resolution applied:
   ws-humantask.xsd
      - simplified tQuery and tExpression complex type definitions (lines 454-468)




[BP-56] Rename ProtocolMsgType

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask-protocol.wsdl

Description: All types in WSHT have a prefix 't' except the ProtocolMsgType in ws-humantask-protocol.wsdl.

Proposal: Rename ProtocolMsgType to tProtocolMessage:
<xsd:complexType name="tProtocolMessage">
   <xsd:sequence>
      <xsd:any maxOccurs="unbounded" minOccurs="0" namespace="##other"
   processContents="lax"/>
      </xsd:sequence>
   <xsd:anyAttribute namespace="##any" processContents="lax"/>
 </xsd:complexType>

and adapt the corresponding references:
<xsd:element name="skipped" type="tProtocolMessage"/>
<xsd:element name="fault" type="tProtocolMessage"/>
<xsd:element name="exit" type="tProtocolMessage"/>

 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:45 AM ]

Issue resolution applied:
   ws-humantask-protocol.wsdl
      - renamed ProtocolMsgType to tProtocolMsgType (lines 14 and 22-24)
   ws-humantask-1.1-spec.doc
      - renamed ProtocolMsgType to tProtocolMsgType (lines 2081, 2093, 2100, 2109)




[BP-55] Import of XML Schema Artifacts

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: WS-HumanTask 1.1, section 2.5.2 "Properties", fourth bullet.

Description: Section 2.5.2 defines the properties of a humanInteractions element. The fourth bullet is about the "import" element and describes imports of other Human Task or WSDL definitions. Parameters of logical people groups reference arbitrary XML Schema types in their "type" attribute. Therefore, XML Schema artifacts must be imported as well.

Proposal: In WS-HumanTask 1.1, section 2.5.2 "Properties", fourth bullet, add XML Schema to the possible import types.

A. Change the sentence in line 400 from:
"The value of the importType attribute MUST be set to http://docs.oasis-open.org/ns/bpel4people/ws-humantask/200803 when importing human interactions definitions, or to http://schemas.xmlsoap.org/wsdl/ when importing WSDL 1.1 documents"
to:
"The value of the importType attribute MUST be set to http://docs.oasis-open.org/ns/bpel4people/ws-humantask/200803 when importing human interactions definitions, to http://schemas.xmlsoap.org/wsdl/ when importing WSDL 1.1 documents, or to http://www.w3.org/2001/XMLSchema when importing XML Schema documents"

B. Change the sentence in line 409 from:
"A WS-HumanTask Definition MUST import all WS-HumanTask and WSDL definitions it uses"
to:
"A WS-HumanTask Definition MUST import all WS-HumanTask, WSDL definitions, and XML Schema definitions it uses".

 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:44 AM ]

Issue resolution applied:
   ws-humantask-1.1-spec.doc
      - added import type xsd to properties in section 2.5.2 (line 404 and 411)




[BP-54] Wrong name in getMyTasksResponse

Created: 02/Dec/08  Updated: 10/Dec/08

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask-api.wsdl

Description: The element name within the getMyTaskResponse element is not in line with the overall naming policy.

Note that this issue is obsolete if the issue "Cleanup tTask Definitions" is resolved as proposed.

Proposal: Rename "taskAbstract":
<xsd:element name="getMyTasksResponse">
  <xsd:complexType>
    <xsd:sequence>
<xsd:element maxOccurs="unbounded" minOccurs="0" name="taskAbstract"
type="htt:tTask"/>
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>

to "task":
<xsd:element name="getMyTasksResponse">
  <xsd:complexType>
    <xsd:sequence>
<xsd:element maxOccurs="unbounded" minOccurs="0" name="task"
type="htt:tTask"/>
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>


 

 Comments 

 

 

Comment by Luc Clement [ 10/Dec/08 10:54 AM ]

This issue was closed with no action.




[BP-53] XML Schema Element Name Resolution in WSDL

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask-api.wsdl, ws-humantask-protocol.wsdl

Description: In both WSDL artifacts, the XML Schema nested inside the <wsdl:types> section has no target namespace attribute. Although this is acceptable from an XML point of view, it seems to provide more clarity if the target namespace for the inlined XML Schema is declared explicitly.

Proposal: Use the WSDL target namespace for the inlined XML Schema *explicitly*, that is, copy the WSDL targetNamespace from the <wsdl:definitions> element to the <xsd:schema> element inside <wsdl:types>.

1. In "ws-humantask-protocol.wsdl", from:

<wsdl:types>
    <xsd:schema blockDefault="#all" elementFormDefault="qualified">

to:

<wsdl:types>
<xsd:schema targetNamespace="http://docs.oasis-
open.org/ns/bpel4people/ws-humantask/protocol/200803"
blockDefault="#all" elementFormDefault="qualified">

2. In "ws-humantask-api.wsdl", from:

<wsdl:types>
    <xsd:schema blockDefault="#all" elementFormDefault="qualified">

to:

<wsdl:types>
<xsd:schema targetNamespace="http://docs.oasis-
open.org/ns/bpel4people/ws-humantask/api/200803"
blockDefault="#all" elementFormDefault="qualified">



 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:15 AM ]

Issue resolution applied:
   ws-humantask-api.wsdl
      - added targetNamespace="http://docs.oasis-open.org/ns/bpel4people/ws-humantask/api/200803" to schema element
   ws-humantask-protocol.wsdl
      - added targetNamespace="http://docs.oasis-open.org/ns/bpel4people/ws-humantask/protocol/200803" to schema element




[BP-52] Rename Priority Expression Type

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask.xsd

Description: The name of the XML Schema complex type for priority expressions in the WS-HumanTask language XML Schema is not consistent with other type names for expressions. Moreover, it has the same local name as the XML Schema simple type used in the WS-HumanTask data types XML Schema (not an XML problem, but confusing).

Proposal: Rename tPriority to tPriority-expr in ws-humantask.xsd (consistent with other *-expr types):
<xsd:complexType name="tPriority-expr" mixed="true">
  <xsd:complexContent mixed="true">
    <xsd:extension base="tExpression" />
  </xsd:complexContent>
</xsd:complexType>

This would also require a change of: <xsd:element name="priority" type="tPriority" /> to
 <xsd:element name="priority" type="tPriority-expr" />

 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:14 AM ]

Issue resolution applied:
   ws-humantask.xsd
      - renamed tPriority to tPriority-expr in ws-humantask.xsd (lines 172 and 469)




[BP-51] Move tTime to ws-humantask-types.xsd

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König

Target: ws-humantask-api.wsdl and ws-humantask-types.xsd

Description: The <types> section of the ws-humantask-api.wsdl only contains request/response wrapper elements used in WSDL messages. The only exception is the tTime element.

Proposal: Move the tTime element definition from ws-humantask-api.wsdl to ws-humantask-types.xsd (note that its namespace will change):
<xsd:complexType name="tTime">
  <xsd:choice>
    <xsd:element name="timePeriod" type="xsd:duration"/>
    <xsd:element name="pointOfTime" type="xsd:dateTime"/>
  </xsd:choice>
</xsd:complexType>

Add the "htt" namespace prefix to the tTask type reference in suspendUntil element in ws-humantask-api.wsdl:
<xsd:element name="suspendUntil">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="identifier" type="xsd:anyURI"/>
      <xsd:element name="time" type="htt:tTime"/>
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>

 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:13 AM ]

Issue resolution applied:
   ws-humantask-api.wsdl
      - removed tTime complex type definition
      - added namespace prefix htt to tTime reference (line 112)
   ws-humantask-types.xsd
      - added tTime complex type definition (lines 183-188)




[BP-50] Refactor User, Group, and OrganizationalEntity

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König - (see: http://lists.oasis-open.org/archives/bpel4people/200812/msg00005.html)

Target: WS-HumanTask 1.1, BPEL4People 1.1, ws-humantask-types.xsd, ws-humantask-context.xsd, ws-humantask-api.wsdl, and all examples

Description: Multiple WS-HT XML Schema artifacts ("ws-humantask-types.xsd", "ws-humantask-context.xsd", "ws-humantask-api.wsdl") import the WS-HT language XML Schema ("ws-humantask.xsd") only to reference the data types User/Group/OrganizationalEntity. However, none of them requires any other element from the WS-HT language XML Schema (note that this would not even be meaningful).

Proposal: Refactor the User/Group/OrganizationalEntity definitions in order to reduce the complexity of import relationships (see diagram - note that the diagram may have been transmitted as attachment).

(see: http://lists.oasis-open.org/archives/bpel4people/200812/msg00005.html)
 

>>> Note: This proposal implies a namespace change for the User/Group/OrganizationalEntity data types which are used in many places.

Summary of changes:
A. Move the User/Group/OrganizationalEntity-related definitions (listed below) into the WS-HT data types XML Schema ("ws-humantask-types.xsd").
B. Remove the import of the WS-HT language XML Schema (in "ws-humantask-types.xsd", "ws-humantask-context.xsd", "ws-humantask-api.wsdl").
C. Change the namespace prefix for User/Group/OrganizationalEntity from "htd:" to "htt:" in the BPEL4People and WS-HumanTask specs as well as in all referencing XML artifacts (including examples).

XML Schema element and type definitions that will be moved into ws-humantask-types.xsd:

  <!-- data types for people assignment -->
  <xsd:element name="user" type="tUser" />
  <xsd:simpleType name="tUser">
    <xsd:restriction base="xsd:string" />
  </xsd:simpleType>
  
  <xsd:element name="group" type="tGroup" />
  <xsd:simpleType name="tGroup">
    <xsd:restriction base="xsd:string" />
  </xsd:simpleType>
  
  <xsd:element name="organizationalEntity" type="tOrganizationalEntity" />
  <xsd:complexType name="tOrganizationalEntity">
    <xsd:choice>
      <xsd:element ref="users" />
      <xsd:element ref="groups" />
    </xsd:choice>
  </xsd:complexType>
  
  <!-- elements and types for organizational entities -->
  <xsd:element name="users" type="tUserlist" />
  <xsd:complexType name="tUserlist">
    <xsd:sequence>
      <xsd:element ref="user" minOccurs="0" maxOccurs="unbounded" />
    </xsd:sequence>
  </xsd:complexType>
  
  <xsd:element name="groups" type="tGrouplist" />
  <xsd:complexType name="tGrouplist">
    <xsd:sequence>
      <xsd:element ref="group" minOccurs="0" maxOccurs="unbounded" />
    </xsd:sequence>
  </xsd:complexType>

 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:13 AM ]

Issue resolution applied:
   ws-humantask.xsd
      - removed elements orgaizationalEntity, user, users, group, groups, and their complex type definitions
   ws-humantask-types.xsd
      - added elements orgaizationalEntity, user, users, group, groups, and their complex type definitions
      - changed namespace prefix of associated element and type references (htd->(no prefix))
      - removed namespace definition xmlns:htd="http://docs.oasis-open.org/ns/bpel4people/ws-humantask/200803"
   ws-humantask-context.xsd
      - changed namespace prefix of associated element and type references (htd->htt)
      - removed namespace definition xmlns:htd="http://docs.oasis-open.org/ns/bpel4people/ws-humantask/200803"
   ws-humantask-api.wsdl
      - changed namespace prefix of associated element and type references (htd->htt)
      - removed namespace definition xmlns:htd="http://docs.oasis-open.org/ns/bpel4people/ws-humantask/200803"
   ws-humantask-example-claim-approval.tel
      - changed namespace prefix of associated element and type references (htd->htt)
      - added namespace definition xmlns:htt="http://docs.oasis-open.org/ns/bpel4people/ws-humantask/types/200803"
      - added htt namespace to xsi:schemaLocation attribute for example validation
   bpel4people-example-employee-of-the-month.bpel
      - changed namespace prefix of associated element and type references (htd->htt)
      - added namespace definition xmlns:htt="http://docs.oasis-open.org/ns/bpel4people/ws-humantask/types/200803"
      - added htd and htt namespaces to xsi:schemaLocation attribute for example validation
   ws-humantask-1.1-spec.doc
      - changed namespace prefix of associated element and type references (htd->htt) (>100x)
   bpel4people-1.1-spec.doc
      - changed namespace prefix of associated element and type references (htd->htt) (>40x)




[BP-49] MaxOccurs Constraint Missing in tFrom

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Dieter König
Target: ws-humantask.xsd

Description: In WS-HumanTask 1.1, section 3.2.1, the "from" element is defined with zero or more arguments - the XML Schema definition only has minOccurs="0" and the missing maxOccurs attribute implies a default value of 1.

Note that the similar problem does not occur in bpel4people.xsd as the "argument" is an extension child element of <bpel:from> which already allows an arbitrary number of extension child elements.

Proposal: In ws-humantask.xsd, add maxOccurs="unbounded" to the "argument" element inside the "tFrom" definition as shown below.

<xsd:complexType name="tFrom" mixed="true">
  <xsd:complexContent>
    <xsd:extension base="tExtensibleMixedContentElements">
      <xsd:sequence>
        <xsd:choice>
          <xsd:element name="argument" type="tArgument" minOccurs="0"
            maxOccurs="unbounded" />
          <xsd:element name="literal" type="tLiteral" minOccurs="0" />
        </xsd:choice>
      </xsd:sequence>
      <xsd:attribute name="expressionLanguage" type="xsd:anyURI" />
      <xsd:attribute name="logicalPeopleGroup" type="xsd:QName" />
    </xsd:extension>
  </xsd:complexContent>
</xsd:complexType>

 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 10:12 AM ]

Issue resolution applied:
   ws-humantask.xsd
      - added maxOccurs="unbounded" to the "argument" element inside the "tFrom" definition (line 422-423)




[BP-48] Abstract BPEL Process Namespace

Created: 02/Dec/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed By: Dieter König
Target: BPEL4People 1.1

Description: Section 7.1.1 mentions a <bpel:opaqueActivity> which is defined in the namespace for abstract BPEL processes - however, the prefix "bpel" is associated with executable BPEL processes.

Proposal: In section "Declared XML Namespace(s)" on page 2, add a new namespace entry for the abstract BPEL namespace:
abstract - http://docs.oasis-open.org/wsbpel/2.0/process/abstract
and use the namespace prefix in section 7.1.1:
<abstract:opaqueActivity>

 

 Comments 

 

 

Comment by Dieter Koenig [ 20/Feb/09 08:00 AM ]

Issue resolution applied:
   bpel4people-1.1-spec.doc
      - added abstract BPEL namespace on page 2
      - changed namespace prefix of opaque activity in 7.1.1




[BP-47] Determining that a people activity's task has been skipped

Created: 30/Oct/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted: Michael Rowley

TARGET: B4P

DESCRIPTION:

There does not appear to be any reliable way for a process definition to check if a previous activity has been skipped. Section 4.8 states: "The people activity goes to final state Obsolete if the task is skipped." However, there is no way for a process to include a condition that checks the state of a people activity.

PROPOSAL:

Add a new XPath extension function called getState, with the following entry in the table of extension functions:

Function: getState

Returns: Returns the current state of the people activity, which will be one of:
    "INACTIVE", "RUNNING", "COMPLETED", "OBSOLETE", "TERMINATED"

Input/Output:
    In: people activity name (xsd:string)
    Out: the current state (xsd:string)


The state transition diagram in section 4.8 should also remove the transitions to the "closed" state. They are currently unlabeled, which appears to mean that the transition to "closed" happens immediately, preventing anyone from ever seeing the states 4 states that precede "closed".

 

 Comments 

 

 

Comment by Dave Ings [ 12/Nov/08 03:25 PM ]

Assigned to Dieter as per 11/12 TC. Ivana also wishes to contribute.

Comment by Dave Ings [ 26/Nov/08 10:34 AM ]

Dieter withdrew his concerns, so Michael's proposal was accepted as per 11/26 TC minutes.

Comment by Dieter Koenig [ 20/Feb/09 07:59 AM ]

Issue resolution applied:
   bpel4people-1.1-spec.doc
      - added getState operation in section 5
      - DID NOT (!) update state diagram in 4.8 (remove transitions to closed state) - to be discussed
      - uploaded Visio diagram source "bpel4people-state-diagram.vsd" into SVN




[BP-46] Inconvenient Return Type of getFault API Operation

Created: 24/Oct/08  Updated: 12/Nov/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Vinkesh Mehta

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Submitted by: Dieter König

TARGET: WS-HumanTask 1.1, section 6.1.1 "Participant Operations" and
"ws-humantask-api.wsdl" / "ws-humantask-types.xsd"

DESCRIPTION: The "getFault" operation returns the fault name and the fault
data for a specified task. Consequently, the XML Schema element definition
for the "getFaultResponse" response wrapper element contains two elements
"faultName" and "faultData". Note that the "getFault" operation is the only
operation that does not return a single output element.

While this is still a correct use of document-literal wrapper style (as
defined in JAX-WS 2.0, section 2.3.1.2 "Wrapper Style"), it is inconvenient
for some "WSDL to Java" mapping tools. If a response wrapper contains a
single wrapper child element then this can be mapped to a Java method
return value (see JAX-WS 2.0, section 2.3.2 "Parameter Order and Return
Type").

PROPOSAL: Wrap the two elements into a single child element such that it
can be mapped to a method return value.

A. In WS-HumanTask 1.1, section 6.1.1 "Participant Operations", table entry
"getFault", column "Parameters", paragraph "Out", replace the two bullets
"faultName" and "faultData" by a single bullet "fault - contains the fault
name and the fault data".

B. In "ws-humantask-api.wsdl", replace the "getFaultResponse" response
wrapper element definition:

<.xsd:element name="getFaultResponse">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="faultName" type="xsd:NCName" />
      <xsd:element name="faultData" type="xsd:anyType" />
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>

by::

<xsd:element name="getFaultResponse">
  <xsd:complexType>
    <xsd:sequence>
      <xsd:element name="fault" type="htt:tFault" />
    </xsd:sequence>
  </xsd:complexType>
</xsd:element>

and add the "tFault" complex type definition to "ws-humantask-types.xsd":

<xsd:complexType name="tFault">
  <xsd:sequence>
    <xsd:element name="faultName" type="xsd:NCName" />
    <xsd:element name="faultData" type="xsd:anyType" />
  </xsd:sequence>
</xsd:complexType>

 

 Comments 

 

 

Comment by Luc Clement [ 29/Oct/08 10:21 AM ]

Approved as resolved during the 29 Oct 08 TC called. The proposal was accepted as is.




[BP-45] Task History

Created: 01/Oct/08  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Relates to

relates to

BP-70

Complement simple query operations

Applied

 

 Description 

 

 

Submitted by: Michael Rowley

TARGET: WS-HT

DESCRIPTION:

Tasks should record the history of actions against them and the task API should have a way to access that history.

PROPOSAL:

http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200904/msg00012.html

 

 Comments 

 

 

Comment by Dave Ings [ 01/Oct/08 02:08 PM ]

Moved from new to open state during 10/1 TC.

Comment by Dave Ings [ 30/Jan/09 03:51 PM ]

As per the 2009-01 F2F, much discussion about this issue (see minutes). Gerhard to open (BP-70) issue to define a consistent paging approach.




[BP-44] Discovering the operations available for the task

Created: 01/Oct/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Relates to

relates to

BP-69

Supporting Batch Operations

Applied

 

 Description 

 

 

Submitted by: Michael Rowley

TARGET: WS-HT

DESCRIPTION:

The operations that are available for a task varies depending on the state that the task is in and the role of the person who might do the operation. The task API should make it possible to ask which operations are currently available for a task for a specific user.

PROPOSAL:

See thread http://lists.oasis-open.org/archives/bpel4people/200812/msg00034.html

 

 Comments 

 

 

Comment by Dave Ings [ 01/Oct/08 02:08 PM ]

Moved from new to open state during 10/1 TC.

Comment by Dave Ings [ 30/Jan/09 03:46 PM ]

As per 2009-01 F2F minutes, Michael will submit a revised proposal that addresses extensibility, and Gerhard will submit an issue (BP-69) that addresses single operations on a set of tasks.

Comment by Dave Ings [ 30/Jan/09 03:56 PM ]

Resolved as per 2009-01 F2F minutes. Revised proposal at:

http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200901/msg00041.html




[BP-43] Permission Matrix should be included in WS-HT Spec

Created: 18/Sep/08  Updated: 06/Jan/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Krasimir Nedkov

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Benjamin Notheis

TARGET: WS-HT

DESCRIPTION:
The information about which Generic Human Roles are allowed to trigger which operations on task instances is conveyed in both the prose definition of Generic Human Roles in section 3.1 and the definition of Participant Operations (section 6.1.1) and Administrative Operations (section 6.1.4).
There should be a matrix somewhere in the spec that is supposed to serve multiple purposes:
- Function as single source of truth regarding permissions, as the information is currently scattered across several locations within the spec
- Provide a comprehensive view on permissions per Generic Human Role and thus the role definition as such
- Facilitate identification of potential glitches in role definitions

PROPOSED RESOLUTION:
Enhance section 6 in the WS-HT spec with a matrix such as the one contained in the attached Excel sheet.
The attached Excel sheet already contains the resolutions of BP-40 that we agreed on during the first day of our F2F.
By including this matrix, the "Authorization" column of the tables in section 6 should be disposed in order to avoid redundant information (and potential inconsistencies).

See attachment to email http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200809/msg00052.html
Attachment: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200809/xls00001.xls

During the Nov 12 telecon we agreed to amend the proposal slightly making the "table become the authoritative source of behaviour"

 

 Comments 

 

 

Comment by Dave Ings [ 01/Oct/08 02:08 PM ]

Moved from new to open state during 10/1 TC.




[BP-42] WS-HT should use consistent naming for datetime properties

Created: 18/Sep/08  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Phillip Allen

TARGET: WS-HT

DESCRIPTION:
(From the F2F)
There are several datetime properties used on various operations or contained within composite data types of the WS-HT spec.
Examples:
AttachedAt
AddedAt
CreatedOn
ActivationTime
ExpirationTime
StartBy
CompleteBy

All of these refer to date time properties; some set by the task processor, some set by the human task. However, there are four different suffixes used to delineate the fact that it is a date time: On, By, At, and Time. This is inconsistent. In particular, 'At' connotes a concept of place as opposed to time.
 

PROPOSED RESOLUTION:
Determine a standard set of suffixes for datetime fields, and then apply throughout the document.

Proposal submitted at http://lists.oasis-open.org/archives/bpel4people/200905/msg00017.html

 

 Comments 

 

 

Comment by Dave Ings [ 01/Oct/08 02:07 PM ]

Moved from new to open state during 10/1 TC.

Comment by Dave Ings [ 30/Jan/09 03:20 PM ]

Assigned to Phil as per 2009-01 F2F minutes.

Comment by Luc Clement [ 27/May/09 10:57 AM ]

Accepted as resolved 27 May 2009

Comment by Ivana Trickovic [ 03/Jun/09 05:34 AM ]

The issue resolution applied in CD-03, Rev3.




[BP-41] Last modification information for tasks

Created: 11/Sep/08  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Michael Rowley

TARGET: WS-HT

DESCRIPTION:

Tasks currently have information about who created them and when. Here are two of the child elements of the <task> element:

    <xsd:element name="createdOn"
                 type="xsd:dateTime"/>
    <xsd:element name="createdBy"
                 type="xsd:string" minOccurs="0"/>

However, there is currently no recorded information about the last modification of a task.

PROPOSED RESOLUTION:

Two new child elements should be added to the <task> element, in order to record by who and when the task was last modified:

    <xsd:element name="lastModifiedTime"
                 type="xsd:dateTime"/>
    <xsd:element name="lastModifiedBy"
                 type="xsd:string" minOccurs="0"/>

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 11:28 AM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 15/Apr/09 11:12 AM ]

Last email on this issues: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200904/msg00016.html




[BP-40] Who can add attachments to unclaimed tasks?

Created: 11/Sep/08  Updated: 12/Nov/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Michael Rowley

TARGET: WS-HT

DESCRIPTION:

Section 6.1.1 says that addAttachment() is only available to actual owners and administrators. However, section 3.1 says that potential owners "can influence the progress of the task, for example by changing the priority of the task, adding ad-hoc attachments or comments."

PROPOSED RESOLUTION:

Allow potential owners to add attachments.

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 11:27 AM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 19/Sep/08 01:24 PM ]

Notes from the Sept FTF:

Michael R: At the F2F, it was noted that this inconsistency also exists for setPriority.

We voted to resolve this issue with the following: Change 6.1.1 to allow potential owners to add attachments and setPriority, but only in the ready state.




[BP-39] Getting or deleting individual attachments

Created: 11/Sep/08  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Michael Rowley

TARGET: WS-HT

DESCRIPTION:

Currently there are operations for getAttachments(name) and deleteAttachment(name). These get or delete _all_ of the attachments of the given name (names are not unique, so there may be several). It should be possible to look at single attachment and possibly delete it, without having to always look at and delete all the attachments that share a name.

This is especially important when you have a UI that presents a list of attachments to the user. The user should be able to click on one and see just that attachment, or delete one, and know that only that attachment will be deleted, even if others share the same name.

PROPOSED RESOLUTION:

Introduce the concept of attachment IDs. Each attachment would have a unique ID within the process. The attachment IDs would be returned as part of the getAttachmentInfo structure.

Introduce two new operations:
getAttachmentByID(attachmentID)
deleteAttachmentByID(attachmentID)

The addAttachment() function also needs to be changed to return an attachment ID.

--- Notes from Sept FTF ---
MichaelR: Note from the F2F on this...

We should remove getAttachments(name) and deleteAttachments(name) and replace them with the functions that take IDs. The names of the functions then don't need "ByID" and can just be called getAttachment(ID) and deleteAttachment(ID).

An open question is whether Attachments should be process scoped or task scoped, given that task infrastructure is probably what will be generating these IDs.
We considered whether there should be an update function, but thought that delete & add might possibly be sufficient.

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 11:27 AM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 29/Apr/09 10:36 AM ]

Resolved during the 29 Apr 09 TC telecon




[BP-38] Comment propagation

Created: 11/Sep/08  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Michael Rowley

TARGET: BPEL4People

DESCRIPTION:

Sometimes the comments that are collected during the execution of one people activity need to be visible to people working on later people activities. This is not currently possible.

It also should be possible to access the comments of a task from XPath expressions in the B4P process. This would make it possible to record the comments in some other document.

PROPOSED RESOLUTION:

Introduce a new child element of a people activity:

<b4p:propagateCommentsFrom>
    <b4p:activityRef>peopleActivityName</b4p:activityRef>*
</b4p:propagateCommentsFrom>

This causes any comments that are present on the named people activities to be copied into the context of the target activity at the time that the target activity becomes ready to execute. If more than one activity in the process has a name referenced from a <b4p:activityRef> element, then all matching activities have their comments propagated. Since comment IDs are unique only within a single task, when comments are propagated they may get new IDs.

We should also add a new XPath function "b4p:getComments(activityName)", which returns the comments for a task using WS-HT's existing tComment type.

To support these changes, the tHumantTaskContext type needs to change to include a <comments> child element, analogous to the <attachments> element.

---- Sep FTF ----

MichaelR: Here are notes from the F2F discussion of this issue:

We talked about having a different approach to sharing comments among people activities. Rather than having to explicitly mark comment propagation flows, it was thought that it is preferable to model sharing by marking a scope with @shareComments=true. This would apply to all activities within the scope and all inner scopes. Inner scopes cannot override sharing, as that would overly complicate the model). However, individual activities can opt out of sharing (perhaps a @dontShareComments attribute).

** see note below date 29 Apr ** Runtimes should allow individual comments to be marked as to whether they should propagate. It should be an enumeration, even if it starts with only two values (private & public)

Comments propagate after the task completes. Propagated comments also propagate metadata (createdBy, lastModifiedBy, etc.)

If getComments is called on a task that is inside an event handler or parallel foreach (where there may be multiple parallel instances), then the result will depend on where the getComments call is made from. If the getComments call is made from within the parallelized scope, then it will return only the comments associated with the scope instance that it is part of. If the getComments call is made from outside of that scope, then it returns a union of all of the comments from every parallel instance of the task.

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 11:27 AM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 29/Apr/09 10:47 AM ]

29 Apr: During the 29 Apr 09 telecon we agreed and approved as resolve the proposal (above) with the need to **delete the paragraph about making individual comments private/public**




[BP-37] Deleting or updating comments

Created: 11/Sep/08  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Michael Rowley

TARGET: WS-HT

DESCRIPTION:

It is currently possible to add comments to a task, but it is not possible to delete those comments. It should be possible to do this. It should also be possible to update a comment.

PROPOSED RESOLUTION:

Add a commentID field to comments which uniquely identifies a comment for a task. The ID only needs to be unique for a single task.

Introduce two new operations on tasks:

  * deleteComment(taskID, commentID)

    Deletes the identified comment.

  * updateComment(taskID, commentID, commentString)

    Replaces the identified comment with the contents of the 'commentString' parameter.

The addComment() function also needs to be changed to return a comment ID.

------ From Sep FTF -----

MichaelR: When this issue was discussed at the F2F, we also agreed to add lastModifiedBy and lastModifiedOn fields to tComment, which is set when updateComment is called.

These two operations can be performed by the same roles as addComment (potential owners, actual owner, business administrator).

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 11:26 AM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 29/Apr/09 10:24 AM ]

Approved as resolved 29 Mar 09




[BP-36] Getting all task details in a single request

Created: 11/Sep/08  Updated: 10/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Micheal Rowley

TARGET: WS-HT

DESCRIPTION:

Currently, the WS-HT interface is fine grained. If an application wants to present the user with all the information about a task on a single screen (other than the contents of the attachments, which are accessed separately), then it would be necessary to call all of the following operations in the current interface:
- getAttachmentInfos
- getComments
- getRenderingTypes
- getRendering
- getTaskInfo
- getTaskDescription
- getInput
- getOutput
- getFault
- getOutcome

PROPOSED RESOLUTION:

We should add an operation called: getTaskDetails which returns all of the information about a task other than the contents of its attachments. This operation takes a task ID as input and returns a union of the contents returned by the operations listed above. Some of the operations above require an additional parameter to get a specific item (getRendering, getTaskDescription, getInput and getOutput). The getTaskDetails operation returns all valid return values of these operations for the given task ID (all renderings, task descriptions, inputs and outputs).

--- Sept FTF Notes ----

MichealR: In yesterday's meeting we informally agreed that this should be handled by a new operation:

getTaskDetails

returns any or all details of a task, except the contents of the attachments.

Parameters:
- taskID:
The id of the task being queried.
- properties:
Optional list of properties of the task that should be provided. Properties are named by the local part of the QName of the element returned for task details.
If it is not specified, then all properties are returned.
If it is specified, then only the properties specified are returned. In the case that multiple elements have the same local part (which can happen for extensions from two different namespaces) then all of the matching properties are returned.
- renderingPreferences:
Optional list of rendering types, in order of preference.
If properties is not specified or includes "renderings" then...
     If renderingPreferences is not specified, the result includes all renderings.
     If it is specified, then the task's rendering whose type is earliest in the provided list of rendering types is returned.

I will be drawing up a more formal proposal.

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 11:26 AM ]

Opened at September 2008 F2F.

Comment by Dave Ings [ 30/Jan/09 03:48 PM ]

As per 2009-01 F2F, this proposal needs to be reconciled with BP-61.

Comment by Luc Clement [ 15/Apr/09 11:11 AM ]

The last tread on this topic is at http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200904/msg00013.html




[BP-35] Value and semantics of suspend/resume

Created: 03/Sep/08  Updated: 12/Nov/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Vinkesh Mehta

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

Proposed by: Michael Rowley

TARGET: WS-HT

DESCRIPTION:

Suspend and resume are currently available to all tasks irrespective of the state they are in. They are also available to the business owner of the task.

It is not clear what the value is of having suspend be available to the owner of a claimed or started task. The specification implies (although it should state it explicitly) that escalations and deadlines continue as scheduled irrespective of the use of suspend. So, why would an owner suspend his own task, especially when the owner is already able to stop and later re-start a task?

More generally, what are the semantics of suspend? They aren't really explained in the current spec, except in relation to the state diagram in section 4.7.

PROPOSAL:

(Actually more of a suggested course of investigation)

Perhaps task suspension should be available only to roles other than the owner or potential owner role (such as business administrators) with a purpose of preventing tasks from being claimed or worked on by owners.

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 11:28 AM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 29/Oct/08 11:27 AM ]

At the 29 Oct we approved as resolved with the following update to (para 2) of http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200810/msg00008.html:

The revised text approved:

"Thus, authorization for suspend/suspendUntil/resume should be restricted to administrators and actual owners, and exclude potential owners, to prevent ordinary users from interfering with the first scenario. We propose to update section 6.1.1 of the spec accordingly."




[BP-34] B4P spec doesn't support error potential owner

Created: 02/Sep/08  Updated: 06/Jan/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

 

 Description 

 

 

Proposed by : Ravi Rangaswamy

TARGET: WS-HT, General

DESCRIPTION: WS-HT should support assigning the task to an error
assignee if there are any errors in determining the potential owners.
Computation of the potential owners of a task as part of people
assignment involves complex operations like evaluating people queries,
XPath expressions, etc. and as such errors might occur and a potential
owner could not be identified. This will cause tasks erroring half way
in complex routing logic

RELATED ISSUES:
• Support for routing patterns and policies
• Support for single routing pattern
• Support for sequential routing pattern
• Support for parallel routing pattern

PROPOSAL: Define an error potential owner is used to assign the task to
when there is an error in task assignment during the evaluation of the
potential owners. Error conditions include invalid task owners and
incorrect XPath expressions in the people assignment. The error assignee
will be able to reassign the task when such event happens.

Syntax

<xsd:complexType name="tOnErrorPotentialOwner">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:element name="from" type="tFrom"/>
</xsd:sequence>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

Example

In the example below, the task will be assigned to the logical people
group errorHandlers if there are any errors during the task assignment.

<htd:peopleAssignments>
...
<htd:onErrorPotentialOwner>
<htd:from logicalPeopleGroup="errorHandlers">
<htd:argument name="region">
htd:getInput("part1")/region
</htd:argument>
</htd:from>
</htd:onErrorPotentialOwner>

</htd:peopleAssignments>

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 05:07 PM ]

Closed at September 2008 F2F, since there is already a mechanism to do this.




[BP-33] B4P spec doesn't support skipping participants in task routing

Created: 02/Sep/08  Updated: 18/Feb/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

Issue Links:

Depends on

depends on

BP-28

B4P spec doesn't address task routing...

Applied

 

 Description 

 

 

Proposed by: Ravi Rangaswamy

TARGET: General

Note that this issue only makes sense in the context of BP-28.

DESCRIPTION: The specification should support a skip condition that allows any of
the patterns to be skipped in the potential owner list. Depending on
some condition, a potential owner could be skipped to work on a task
although the user is part of the routing pattern definition.

RELATED ISSUES:
 • Support for routing patterns and policies
 • Support for single routing pattern
 • Support for sequential routing pattern
 • Support for parallel routing pattern

PROPOSAL: Each task routing pattern (single, sequential or parallel)
might have an optional skip condition. If this condition evaluates to
true, the potential owner will be skipped.

Syntax

The syntax for skip condition is shown in the tSequential and tParallel
schema definitions of the task routing pattern proposal.

Example

In the example below, the sequential pattern will be skipped if the
claim account is less than 50000.

  <htd:potentialOwners>
    <htd:sequential name="Claim Processing Agents"
skipCondition="htd:getInput("ClaimApprovalRequest")/amount < 50000">
      <htd:list>
        
<htd:from>htd:getInput("ClaimApprovalRequest")/claimProcessingAgents</htd:from>
      </htd:list>
    </htd:sequential>
    <htd:single name="Claim Processing Review">
      <htd:from>
        <htd:literal>
          <htd:organizationalEntity>
            <htd:users>
              <htd:user>achrist</htd:user>
            </htd:users>
          </htd:organizationalEntity>
        </htd:literal>
      </htd:from>
    </htd:single>
</htd:potentialOwners>

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 05:06 PM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 18/Feb/09 10:57 AM ]

This is being closed with no action. These will be tracked as part of issue BP-28




[BP-32] B4P spec doesn't support early completion of task routing

Created: 02/Sep/08  Updated: 18/Feb/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

Issue Links:

Depends on

depends on

BP-28

B4P spec doesn't address task routing...

Applied

 

 Description 

 

 

Proposed by : Ravi Rangaswamy

TARGET: General

DESCRIPTION: The specification should support specifying early completion of task routing.

Note that this issue only makes sense in the context of BP-28.

Use Case: Consider a complex routing condition involves task going
sequential across multiple assignees. If any of the potential owners
rejects the task, then the task routing should be terminated.

RELATED ISSUES:
• Support for routing patterns and policies
• Support for single routing pattern
• Support for sequential routing pattern
• Support for parallel routing pattern

PROPOSAL: Early completion can be specified in people assignment to
control early completion of the task routing. The early completion is
specified as an XPath expression and if this XPath expression evaluates
to true, then the routing will stop. The early completion condition will
be evaluated when the current task owner completes the task and the next
task assignee is determined.

Example
In the example below, the task routing will stop when the outcome is REJECT.

<htd:peopleAssignments>
...
<htd:earlyCompletion>htd:getOutcome()='REJECT'<htd:earlyCompletion>

<htd:peopleAssignments>

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 05:05 PM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 18/Feb/09 10:56 AM ]

This is being closed with no action. These will be tracked as part of issue BP-28




[BP-31] B4P spec doesn't support parallel routing pattern

Created: 02/Sep/08  Updated: 18/Feb/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

Issue Links:

Depends on

depends on

BP-28

B4P spec doesn't address task routing...

Applied

 

 Description 

 

 

Proposed by: Ravi Rangaswamy

TARGET: General

DESCRIPTION: The specification should support parallel routing pattern. In the
current spec, people assignments are singular in notion. The
potential owners in the sequential pattern can either be done through
logical people groups, literals or an expression.

Use Case: Lets consider a document review process where people from
different departments need to review a document and provide input for a
task. The specification should support that model where multiple people work on a
single task in parallel in a collaborative environment.

RELATED ISSUES:
• Support for routing patterns and policies
• Support for single routing pattern
• Support for sequential routing pattern

PROPOSAL: Provide a mechanism to assign the task in parallel to
users/groups by introducing a parallel routing pattern. In the parallel,
all of the parallel users will work on the task in parallel. Completion
criteria should also be specified to determine the completion of the
parallel tasks and final outcome of the task.

The potential owners in parallel pattern is specified by a htd:list or
htd:branch. The htd:list is specified using a list of htd:from elements
or a management chain. The htd:list element was introduced in the
sequential pattern issue. htd:branch allows modeling complex routing
inside any of the parallel branches. htd:branch contains htd:single,
htd:sequential and/or htd:parallel.

Syntax

<xsd:complexType name="tParallel">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:choice minOccurs="1" maxOccurs="1">
<xsd:element name="firstResponder"
type="tFirResponder"/>
<xsd:element name="allResponders"
type="tParallelCompletionCriteria"/>
<xsd:element name="vote"
type="tParallelCompletionCriteria"/>
</xsd:choice>
<xsd:element name="list" type="tList"/>
<xsd:element name="branch" type="tPotentialOwner"
minOccurs="0"
maxOccurs="unbounded"/>
</xsd:sequence>
<xsd:attribute name="name"
type="xsd:string" use="optional"/>
<xsd:attribute name="skipCondition"
type="xsd:string" use="optional"/>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

<xsd:complexType name="tList">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:choice maxOccurs="1">
<xsd:element name="from" type="tFrom"
minOccurs="0"
maxOccurs="unbounded"/>
<xsd:element name="managementChain"
type="tManagementChain"
minOccurs="0" maxOccurs="1"/>
</xsd:choice maxOccurs="1">
</xsd:sequence>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

<xsd:complexType name="tManagementChain">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:element name="from" type="tFrom"
minOccurs="0"
maxOccurs="unbounded"/>
<xsd:element name="levels" type="xsd:string"
minOccurs="1"/>
<xsd:element name="title" type="xsd:string"
minOccurs="0"/>
</xsd:sequence>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

<!-- tPotentialOwners duplicated from pattern and policy issues doc -->
<xsd:complexType name="tPotentialOwners">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:choice maxOccurs="unbounded">
<xsd:element name="single" type="tSingle"/>
<xsd:element name="sequential"
type="tSequential"/>
<xsd:element name="parallel" type="tParallel"/>
</xsd:choice>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

<xsd:complexType name="tParallelCompletionCriteria">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:element name="defaultOutcome"
type="xsd:string"
minOccurs="1" maxOccurs="1"/>
<xsd:element name="percentageOfOutcome"
type="xsd:string"
minOccurs="1" maxOccurs="1"/>
</xsd:sequence>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

<xsd:complexType name="tFirstResponder">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence></xsd:sequence>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>


Example

In the example below, assuming that the task input ClaimApprovalRequest
has 3 claimProcessingAgents children - jlondon, achrist, jcooper, there
are 3 parallel task each assigned to those 3 users

<htd:parallel name="Claim Processing Review">
<htd:vote>
<htd:defaultOutcome>
string('APPROVE')</htd:defaultOutcome>
<htd:percentageOfOutcome>
number(50)</htd:percentageOfOutcome>
</htd:vote>
<htd:list>
<htd:from>
htd:getInput("ClaimApprovalRequest")/claimAgent
</htd:from>
</htd:list>
</htd:parallel>

In the example below, there are 2 parallel branches. In each branch,
there is a sequential pattern.

<htd:parallel name="Claim Processing Review">
<htd:vote>
<htd:defaultOutcome>
string('APPROVE')</htd:defaultOutcome>
<htd:percentageOfOutcome>
number(50)</htd:percentageOfOutcome>
</htd:vote>
<htd:branch>

<htd:sequential name="Claim Processing Agents">
<htd:list>
<htd:from>
htd:getInput("ClaimApprovalRequest")/claimProcessingAgents
</htd:from>
</htd:list>
</htd:sequential>
</htd:branch>
<htd:branch>
<htd:sequential name="Claim Processing Review Board">
<htd:list>
<htd:from>
<htd:literal>
<htd:organizationalEntity>
<htd:users>
<htd:user>achrist</htd:user>
<htd:user>bpalmer</htd:user>
</htd:users>
</htd:organizationalEntity>
</htd:literal>
</htd:from>
</htd:list>
</htd:sequential>
</htd:branch>
</htd:parallel>

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 05:05 PM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 18/Feb/09 10:56 AM ]

This is being closed with no action. These will be tracked as part of issue BP-28




[BP-30] B4P spec doesn't support sequential routing pattern

Created: 02/Sep/08  Updated: 18/Feb/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

Issue Links:

Depends on

depends on

BP-28

B4P spec doesn't address task routing...

Applied

 

 Description 

 

 

Proposed By: Ravi Rangaswamy

TARGET: General

DESCRIPTION: The specification should support sequential routing pattern. In the
current spec, people assignments are singular in notion. The
potential owners in the sequential pattern can either be done through
logical people groups, literals or an expression.

Use Case: Consider an approval process for OnBoarding where a task has
to be approved by following a management chain. Only if one user
completes the task, it is assigned to the next in the chain. The specification should support the assignment of tasks to a sequence of users/groups.

RELATED ISSUES:
• Support for routing patterns and policies
• Support for single routing pattern
• Support for parallel routing pattern

PROPOSAL: Provide a mechanism to assign the task sequentially to
users/groups by introducing a sequential routing pattern. In the
sequential pattern, the task is assigned to either a list of users or
groups. In each assignment, if there are multiple potential owners than
one can claim the task and become the actual owner. After the actual
user completes the task, it is assigned to the next users/groups in the
sequential pattern.

The sequential routing pattern can either represent a list of sequential
users/groups or a management chain.

In the list of users/groups case, the users/groups to approve the task
are specified in the <htd:from> element. After each task owner completes
the task, the task will be assigned to next user(s)/group(s) in the
list. The order of the htd:from children will be preserved and used in
the ordering of the user/groups evaluated from the list of htd:from
elements.

Management chain indicates a sequence of task assignment that is a
result of computing the management chain in the associated user
directory. The management chain contains the optional first task
assignment. This task assignment could result in one or more potential
owners. If there are multiple potential owners, one user from the list
of potential owners can claim the task and become the actual owner. When
the actual owner completes the task, the actual owner is used as the
base user to compute the management chain. If the first task assignment
is not present, the user who previously completed the task will be used
to compute the management chain. When the first task assignment is
specified in the management chain, optionally the actual owner could
also be specified to indicate who should be the default actual owner in
case there is more than one user in the potential owners list. The
management chain list is controlled by the 2 parameters:
1. Levels: The number of levels in the management chain to actually
assign the task to. The count of level starts from the first assignment
resulting from the evaluation of the management chain. This is an
optional configuration.
2. Highest Title of Approver: The title of the last (highest) user in
the management chain. This is an optional configuration.

Syntax

<xsd:complexType name="tSequential">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:element name="list" type="tList"/>
</xsd:sequence>
<xsd:attribute name="name" type="xsd:string"
use="optional"/>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

<xsd:complexType name="tList">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:choice maxOccurs="1">
<xsd:element name="from" type="tFrom"
minOccurs="0"
maxOccurs="unbounded"/>
<xsd:element name="managementChain"
type="tManagementChain"
minOccurs="0" maxOccurs="1"/>
</xsd:choice maxOccurs="1">
</xsd:sequence>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

<xsd:complexType name="tManagementChain">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:element name="from" type="tFrom"
minOccurs="0"
maxOccurs="unbounded"/>
<xsd:element name="levels" type="xsd:string"
minOccurs="1"/>
<xsd:element name="title" type="xsd:string"
minOccurs="0"/>
</xsd:sequence>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

Examples

In the example below, assuming that the task input ClaimApprovalRequest
has 3 claimProcessingAgents children - jlondon, jstein, wfaulk, the task
will first be assigned to jlondon. After jlondon completes the task, it
will be assigned to jstein. After jstein completes the task, it will be
assigned to wfaulk.

<htd:sequential name="Claim Processing Agents">
<htd:list>
<htd:from>
htd:getInput("ClaimApprovalRequest")/claimProcessingAgents
</htd:from>
</htd:list>
</htd:sequential>

In the example below, the task will first be assigned to user achrist.
After achrist completes the task, it will be assigned to achrist's
manager jlondon. After jlondon completes the task, it will be assigned
to his manager jstein. The management chain will stop at that level
because it has gone 3 levels. If jlondon where the 'Vice President', the
management chain will stop with jlondon.

<htd:sequential name="Claim Processing Review">
<htd:list>
<htd:managementChain>
<htd:from>
<htd:literal>
<htd:organizationalEntity>
<htd:users>
<htd:user>achrist</htd:user>
</htd:users>
</htd:organizationalEntity>
</htd:literal>
</htd:from>
<htd:levels>number(3)</htd:levels>
<htd:title>string('Vice President')</htd:title>
</htd:managementChain
</htd:list>
</htd:sequential>


 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 05:04 PM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 18/Feb/09 10:56 AM ]

This is being closed with no action. These will be tracked as part of issue BP-28




[BP-29] B4P spec doesn't support single approval routing pattern

Created: 02/Sep/08  Updated: 18/Feb/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Unassigned

Resolution:

No Action

Votes:

0

 

Issue Links:

Depends on

depends on

BP-28

B4P spec doesn't address task routing...

Applied

 

 Description 

 

 

Proposed by: Ravi Rangaswamy

TARGET: General

DESCRIPTION: The specification should support a single routing pattern aligned with
other routing patterns. In the current spec, people assignments
are singular in notion, but it doesn't fit with any overall support for
patterns. The potential owners in the single pattern can either be done
through logical people groups, literals or an expression.

RELATED ISSUES:
• Support for routing patterns and policies
• Support for sequential routing pattern
• Support for parallel routing pattern

PROPOSAL: Provide a mechanism to assign the task singularly to
users/groups by introducing a single routing pattern. If there are
multiple potential owners than one can claim the task and become the
actual owner. After the actual user completes the task, the single
pattern completes.

Syntax

<xsd:complexType name="tSingle">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:sequence>
<xsd:element name="from" type="tFrom"
minOccurs="0"
maxOccurs="unbounded"/>
</xsd:sequence>
<xsd:attribute name="name" type="xsd:string"
use="optional"/>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>


Examples

In the example below, assuming that the task input ClaimApprovalRequest
has 3 claimProcessingAgents children - jlondon, jstein, wfaulk, the task
will be assigned to jlondon, jstein and wfaulk. One of them acquires the
task and then the task completes.

<htd:single name="Claim Processing Agents">
<htd:from>
htd:getInput("ClaimApprovalRequest")/claimProcessingAgents
</htd:from>
</htd:single>

In the example below, the task will be assigned to user achrist.

<htd:single name="Claim Processing Review">
<htd:from>
<htd:literal>
<htd:organizationalEntity>
<htd:users>
<htd:user>achrist</htd:user>
</htd:users>
</htd:organizationalEntity>
</htd:literal>
</htd:from>
</htd:single>

 

 Comments 

 

 

Comment by Dave Ings [ 18/Sep/08 05:03 PM ]

Opened at September 2008 F2F.

Comment by Luc Clement [ 18/Feb/09 10:56 AM ]

This is being closed with no action. These will be tracked as part of issue BP-28




[BP-28] B4P spec doesn't address task routing patterns and policies

Created: 02/Sep/08  Updated: 25/Jun/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

File Attachments:

Description: Microsoft WordPatterns-in-ws-humantask-on-cd02.doc     Description: Microsoft WordPatterns-in-ws-humantask-on-cd02_merge.doc     Description: Microsoft ExcelSubtasks_RoutingPatterns_OpenIssues.xls     Description: Microsoft ExcelSubtasks_RoutingPatterns_OpenIssues_090520.xls     Description: Filews-humantask.xsd     Description: Microsoft Wordwsht_composite_tasks.doc    

Issue Links:

Depends on

is depended on by

BP-29

B4P spec doesn't support single appro...

Closed

is depended on by

BP-30

B4P spec doesn't support sequential r...

Closed

is depended on by

BP-31

B4P spec doesn't support parallel rou...

Closed

is depended on by

BP-32

B4P spec doesn't support early comple...

Closed

is depended on by

BP-33

B4P spec doesn't support skipping par...

Closed

 

 Description 

 

 

Proposed by: Ravi Rangaswamy

TARGET: General

DESCRIPTION:

1. The specification should support common patterns for assigning people. In the
current spec, people assignments are singular in notion. The
assignment to potential owners of a task can either be done through
logical people groups, literals or an expression.

However there is no mechanism to specify that a group of people might
work on a task in parallel or in a well-defined sequence or in any
combination of them, the so-called 'task routing pattern'.

In the current model, task routing patterns have to be modeled in the
business process itself. However that makes it hard to manage the task
in its entire context.

2. The specification should support a standard set of 'task routing policies' that
would allow the task modeler to apply certain policies on task routing.
The policies include early routing completion, skipping modeled routing
patterns and error assignment

For example, assume that a task is assigned to a group of people to work
on the task in parallel. One might want to complete all routing as soon
as a certain condition on the task payload evaluates to true.

RELATED ISSUES:
• Support for single routing pattern
• Support for sequential routing pattern
• Support for parallel routing pattern
• Support for early completion
• Support for skip condition
• Support for error assignee

PROPOSAL:

1. Provide a mechanism for people assignments to support common task
routing patterns for Single, Sequence, Parallel and any combination of
them. This is to enable collaborative work on tasks in a well-defined order.

2. Define a set of standard task routing policies that would allow the
task modeler to set the state of a task according to policies evaluated
at runtime.

Detailed proposals are available as part of the associated issues.

Syntax

<xsd:element name="potentialOwners"
type="tPotentialOwners"/>

<!-- MODIFIED: Potential Owners to include chain and groupVote -->
<xsd:complexType name="tPotentialOwners">
<xsd:complexContent>
<xsd:extension base="tExtensibleElements">
<xsd:choice maxOccurs="unbounded">
<xsd:element name="single" type="tSingle"/>
<xsd:element name="sequential" type="tSequential"/>
<xsd:element name="parallel" type="tParallel"/>
</xsd:choice>
</xsd:extension>
</xsd:complexContent>
</xsd:complexType>

<!-- All the other types here are discussed in detail in related issues
doc -->

 

 Comments 

 

 

Comment by Luc Clement [ 02/Sep/08 07:48 PM ]

This Issues was accidentally moved from New to Open.

As of 2 Sep 2008, this issue is New and not yet accepted as Open.

Comment by Dave Ings [ 18/Sep/08 05:03 PM ]

(Formally) opened at September 2008 F2F.

Comment by Luc Clement [ 04/Mar/09 10:56 AM ]

Please review input from http://www.osoa.org/jira/browse/BP-74 that we agreed to handle as part of BP-28

Comment by Hannah Petereit [ 21/Apr/09 08:42 AM ]

Merge document for routing patterns & composite task concepts.

Comment by Hannah Petereit [ 19/May/09 07:31 AM ]

Attached "Open Issues" document to collect all conceptual issues that need to be resolved by the upcoming F2F.

Comment by Hannah Petereit [ 20/May/09 11:28 AM ]

Updated open issues list after 20-May working session -> basis for creation of official TC issues




[BP-27] Test is test issue 2.

Created: 22/Aug/08  Updated: 06/Jan/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Major

Reporter:

Dave Ings

Assigned To:

Unassigned

Resolution:

Fixed

Votes:

0

 

 

 Comments 

 

 

Comment by Dave Ings [ 22/Aug/08 10:49 AM ]

This test issue is no longer required.




[BP-26] This is a test issue.

Created: 22/Aug/08  Updated: 06/Jan/09

Status:

Closed

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Major

Reporter:

Dave Ings

Assigned To:

Unassigned

Resolution:

Fixed

Votes:

0

 

 

 Comments 

 

 

Comment by Dave Ings [ 22/Aug/08 10:49 AM ]

This test issue is no longer required.




[BP-25] WS-HumanTask Attachment Info Data Type element contentType is ambiguous

Created: 25/Jun/08  Updated: 29/Oct/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Gerhard Pfau
TARGET: BPEL4People Working Draft 02 and WS-HumanTask Working Draft 02

DESCRIPTION: The WS-HumanTask Attachment Info Data Type has an element contentType, which is defined in WS-HumanTask, section 3.4.3.1 Ad-hoc Attachments, line 728f as follows: "The contentType of an attachment can be any valid XML schema type, including xsd:any [Note: should say xsd:anyType instead of xsd:any] , or any MIME type." The issue with this definition is that it is not clear if what is specified by contentType is actually a MIME type, an XML schema type, or something else.

PROPOSAL: Define an additional element contentCategory that specifies the category used by contentType. Also define an extensible set of default values for contentCategory including "MIME" and "XSD-Schema".

 

 Comments 

 

 

Comment by Luc Clement [ 16/Jul/08 12:17 PM ]

Proposal Assigned to Gerhard Pfau, 20081709

Comment by Luc Clement [ 20/Aug/08 10:44 AM ]

Approved as resolved 20080820 as proposed at http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200807/msg00019.html

Comment by Dieter Koenig [ 24/Sep/08 11:37 AM ]

Applied BP-25 to WS-HT section 3.4.3.1 and ws-humantask-types.xsd
See SVN revision 54




[BP-24] getOutcome function missing in B4P spec

Created: 19/Jun/08  Updated: 11/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ralf Mueller

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Relates to

relates to

BP-14

Need to provide access to the Outcom...

Applied

 

 Description 

 

 

SUBMITTED BY: Ralf Mueller
TARGET: BPEL4People Specification, Section 5.

DESCRIPTION: It would be valuable for a people activity to query the
task outcome of another people activity

PROPOSAL: Add a new XPath function getOutcome() in the BPEL4People spec, section 5:
Operation Name: getOutcome
Description: Returns the task outcome of the task associated with the people activity.

Parameters: In: people activity name (xsd:string),
                Out: the task outcome (xsd:string)

 

 Comments 

 

 

Comment by Luc Clement [ 16/Jul/08 12:19 PM ]

Proposal Assigned to Ralf Mueller, 20081709

Comment by Luc Clement [ 01/Oct/08 10:37 AM ]

See expanded proposal at: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200809/msg00015.html

During the 1 Oct telecon, we agree to add the following sentence after the Parameters section

Additional sentence: "If the outcome is not present, it returns an empty nodeset."




[BP-23] (Re-)Evaluation and instantiation of logical people groups

Created: 10/Jun/08  Updated: 29/Oct/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

is depended on by

BP-15

What is the purpose of the optional t...

Applied

is depended on by

BP-6

Does ht:getLogicalPeopleGroup() cause...

Applied

 

 Description 

 

 

SUBMITTED BY: Matthias Kloppmann
TARGET: BPEL4People Working Draft 02

DESCRIPTION: When are logical people groups re-evaluated, in particular for parameterized logical people groups.

Resolution of logical people groups happens on several occasions: When referenced from an htd:from clause, or when referenced from a getLogicalPeopleGroup XPath function. When accessed the first time through either means, a logical people group is evaluated. In a BPEL4People process, the same logical people group may be referenced again, possibly with a different set of arguments values. The specification needs to state what the rules are for re-evaluating (re-resolving) the logical people group.

PROPOSAL: Instances of an LPG are per set of values for the parameters. Consequently, when arguments of parameters are changed between referencing an LPG the first time and the second time, it MUST be re-evaluated/re-resolved. When parameters are not changed between referencing an LPG the first time and the second time, it MAY be re-evaluated/re-resolved.
(Motivation: Allow a modeler to reuse a logical people group, e.g., in a loop, with different bindings -- the example we discussed was regionalSalesPeople("Wales"), regionalSalesPeople("Sussex") -- you want to be sure the second one does not retrieve the former result, unless they are really the same. Allow in implementation for caching: When regionalSalesPeople("Wales") is called several times, the result could be cached, because re-resolution is not mandated.)

 

 Comments 

 

 

Comment by Luc Clement [ 16/Jul/08 12:20 PM ]

Proposal Assigned to Matthias Kloppmann, 20081709

Comment by Dave Ings [ 18/Sep/08 05:15 PM ]

Resolved at September 2008 F2F, see 9/18 minutes for details.

Comment by Luc Clement [ 22/Sep/08 08:46 AM ]

Per minutes of the Sept 2008: Proposed motion: accept proposal for BP-23 as
documented here http://www.oasisopen.org/apps/org/workgroup/bpel4people/email/archives/2008
07/msg00012.html




[BP-22] Make b4p:peopleAssignments optional

Created: 10/Jun/08  Updated: 28/Jun/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: B4P, Section 3.1.1

DESCRIPTION: The b4p:peopleAssignments element should be optional.

PROPOSAL: Add a question mark on line 375 and 384 of the b4p spec. For
example:

<b4p:peopleAssignments>?

  <htd:genericHumanRole>+
    <htd:from>...</htd:from>
  </htd:genericHumanRole>

<b4p:peopleAssignments>

 

 Comments 

 

 

Comment by Dave Ings [ 11/Jun/08 03:32 AM ]

Dave Ings: Proposed resolution: Schema needs to be updated accordingly. The changes are required to be made to lines 292, 375, and 384. At the end of 322 add: This element is optional. Before line 399 add a new line saying:If the element peopleAssignments is present it MUST specify assignment for at least one process-related generic human role.

Gerhard Pfau: Dieter: Motion to resolve BP-22 as per the proposed tresolution above. Ralf seconds.
Accepted by unanimous consent

Comment by Dieter Koenig [ 28/Jun/08 08:37 AM ]

Resolution incorporated into the BPEL4People specification, see SourgeForge SVN, revision 18




[BP-21] Precedence of spec vs schema files

Created: 10/Jun/08  Updated: 28/Jun/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Martin Chapman
TARGET: BPEL4People and WS-HumanTask Working

DESCRIPTION: Both specifications contain fragments of schemas and pseudo schemas through the text of the documents, in addition to schemas (and wsdls) in external files. Both specifications must state what is the order of precedence if there is are any discrepancy between the documents and the external files.


PROPOSAL: Add the following text to each document: "Throughout this specification, [WSDL and] schema elements may be used for illustrative or convenience purposes. However, in a situation where those elements within this document differ from the separate [B4P/HT] [WSDL or] schema files, it is those files that have precedence and not this document.

 

 Comments 

 

 

Comment by Dave Ings [ 11/Jun/08 03:31 AM ]

Dave Ings: "Throughout this specification, [WSDL and] schema elements may be used for illustrative or convenience purposes. However, in a situation where those elements or other text within this document contradict the separate [B4P/HT] [WSDL or] schema files, it is those files that have precedence and not this document.

Dave Ings: Suggestion (subject to editor's discretion) is that this text be in section of each spec.

Dave Ings: section two that is

Gerhard Pfau: Martin: Motion to resolve BP-21 as per the text above. Dieter seconds.

Gerhard Pfau: Accepted by unanimous consent.

Comment by Dieter Koenig [ 28/Jun/08 08:38 AM ]

Resolution incorporated into the BPEL4People and WS-HumanTask specifications, see SourgeForge SVN, revision 18




[BP-20] Extensibility of BPEL4People and WS-HumanTask

Created: 10/Jun/08  Updated: 21/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

is depended on by

BP-12

Value of genericHumanRole in setGener...

Applied

 

 Description 

 

 

SUBMITTED BY: Dieter Koenig
TARGET: BPEL4People Working Draft 02 and WS-HumanTask Working Draft 02

DESCRIPTION: Both specs need to clarify the extensibility of the respective
languages. As an example, it should be defined how to introduce new generic
human roles for human tasks (standalone or in the context of BPEL process)
- the current version of the B4P/HTD XML Schema definitions only define a
fixed list of generic human roles.

This new issue is an outcome of a discussion of issue 12
(http://www.osoa.org/jira/browse/BP-12) which is about the possible values
for generic human role parameters on API operations. The resolution of
issue 12 is therefore dependent on a resolution of the general
extensibility addressed in this issue.

PROPOSAL:

 

 Comments 

 

 

Comment by Dave Ings [ 26/Nov/08 10:14 AM ]

For some reason that had not been assigned to Dieter.

Comment by Dave Ings [ 26/Nov/08 10:33 AM ]

Dieter's revised proposal accepted as per 11/26 minutes

Comment by Dieter Koenig [ 20/Feb/09 07:58 AM ]

Issue resolution applied:
   ws-humantask-types.xsd
      - changed simple type tStatus to xsd:string (lines 157-159)
      - added simple type tPredefinedStatus with enumeration (lines 161-174)
   ws-humantask-1.1-spec.doc
      - added pre-states and post-states to table in 6.1.1
      - added normative language about pre-states and post-states in 6.1.1
      - updated state diagram in 4.7 (activate->activation, revoke->release)
      - uploaded Visio diagram source "ws-humantask-state-diagram.vsd" into SVN




[BP-19] Using process variables in constellations 1 and 2

Created: 10/Jun/08  Updated: 29/Oct/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: B4P, Section 4

DESCRIPTION: Constellations 1 and 2 allow the task definition to reference
variables from the enclosing BPEL process. In some cases, these variables
can be used within a part of the task that is not immediately evaluated when
the people activity executes.

For example:

<ht:escalation ...>
   <ht:condition>ht:getActualOwner() = 'Bob' and $v1 = 100</ht:condition>
   ...
</ht:escalation>

Will a change to v1 after the people activity executes but prior to its
escalation having been reached be visible to the task? The spec should say
something about this since it could be an interop issue.

COMMENTARY: I don't have a good use case for this issue since I would expect
the task to make its decisions based on its input data and not the state of
the enclosing process.


PROPOSAL:

The following proposal was approved during the 20080806 telecon by the TC: http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200807/msg00016.html

 

 Comments 

 

 

Comment by Luc Clement [ 16/Jul/08 12:18 PM ]

Proposal Assigned to Phil Allen, 20081709




[BP-18] Task Priority is underspecified

Created: 06/Jun/08  Updated: 30/Jul/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Krasimir Nedkov

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Krasimir Nedkov
TARGET: WS-HT, General

Description:
The task priority is a user-centric property and therefore should be displayable in a human-readable form. Task information is rendered by various applications (task-list-UIs, task-description-Uis, task-UIs), which are decoupled and may even have different vendors. These applications need a common convention how to interpret the priority and how to render it to the end user. Currently the WS-HT specification defines it too loosely just as an integer, which limits the interoperability. Imagine that your applicatiosn receives the value 51435 - what would you display in the UI?

"priority: This element is used to specify the priority of the task. It is an optional element which value is an integer expression. If not present, the priority of the task is unspecified. 0 is the highest priority, larger numbers identify lower priorities. The result of the expression evaluation is of type xsd:integer. The expressionLanguage attribute specifies the language used in the expression. The attribute is optional. If not specified, the default language as inherited from the closest enclosing element that specifies the attribute is used

There is also ambiguity in the current specification about the optional nature of the priority - what is the query behaviour for tasks without priority and what integer correspond to the 'unspecified' value (this should be returned by the getTaskPriority operation).

Proposal:

1. Define an enumeration of 4 fixed integer values and define them as Very High, High, Normal, Low. These MUST be understood by compatible clients. As it would be good to re-use an existing standard, I propose to take the codes from the UN/EDIFACT task priority definition (http://www.unece.org/trade/untdid/d07b/tred/tred4037.htm) and probably extend them. Additional extension would be introducing code 7 with value Low.
2. Define the priority as a mandatory property with default value Normal.
Detailed technical proposal will follow if the issue is opened.

 

 Comments 

 

 

Comment by Dave Ings [ 11/Jun/08 04:23 AM ]

Dave Ings: Proposed resolution of issue 18: Define a range of 0..10 for the task priority with no semantic implied by the priority other than 0 is highest, 10 is lowest with 5 the default.

Dave Ings: (Note: the editors will need to determine all the spec references that need to be revised.)

Gerhard Pfau: Krasimir: Motion to resolve BP-18 as per the proposed resolution above. Matthias seconds.
Accepted by unanimous consent.




[BP-17] Define Conformance Targets

Created: 05/Jun/08  Updated: 11/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ralf Mueller

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Martin Chapman
TITLE: Define Conformance Targets

TARGET: BPEL4People and HumanTask

DESCRIPTION: In order to write a normative statement or conformance clause using RFC2119 language, the target or subject of the rule needs to be included. Some of these are obvious, as we have rules governing schema/document formation. Other targets are less so, especially when it comes to conformance, as we need to identify artefacts that vendors can claim they handle according to the rules of the specs. We should enumerate a list of conformance targets early on, so that every time we use a MUST, MAY, or SHOULD, we know to what the rule applies to.

PROPOSAL: Look at WS-BPEL and SCA specs for inspiration.

 

 Comments 

 

 

Comment by Dave Ings [ 10/Jun/08 07:45 AM ]

Dave Ings: Proposed motion: Add the draft conformance target text to each specification and then replace existing references to "compliant implementations" with references to the new definitions.

Dave Ings: WS-HumanTask Definition any artifact that complies with the human interaction schema and additional constraints defined in this document.

WS-HumanTask Processor any implementation that accepts a WS-HumanTask definition and executes the semantics in this document.

BPEL4People Definition is a WS-BPEL 2.0 process definition that uses the BPEL4People extensions to WS-BPEL 2.0 specified in this document.

BPEL4People Processor any implementation that accepts a BPEL4People definition and executes the semantics defined in this document.

Comment by Dave Ings [ 18/Sep/08 05:19 PM ]

Previously approved (resolved) proposal amended with two new conformance points, at September 2008 F2F, see 9/18 minutes.

BPEL4People spec:
================
- BPEL4People Definition
   BPEL4People Definition is a WS-BPEL 2.0 process definition that
uses the BPEL4People extensions to
   WS-BPEL 2.0 specified in this document

- BPEL4People Processor
   A BPEL4People Processor is any implementation that accepts a
BPEL4People definition and executes the
   semantics defined in this document


WS-HT spec:
==========
- WS-HumanTask Definition
   A WS-HumanTask Definition is any artifact that complies with the
human interaction schema and additional
   constraints defined in this document

- WS-HumanTask Processor
   A WS-HumanTask Processor is any implementation that accepts a WS-
HumanTask definition and executes
   the semantics in this document

- WS-HumanTask Parent (new)
   A WS-HumanTask Parent is any implementation that supports the
Interoperable Protocol for Advanced
   Interactions with Human Tasks as defined in this document

- WS-HumanTask Client (new)
   A WS-HumanTask Client is any implementation that uses the
Programming Interfaces of the WS-HumanTask
   Processor.




[BP-16] Revise Specs to use RFC2119 Keywords

Created: 05/Jun/08  Updated: 18/Feb/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

File Attachments:

Description: Microsoft Wordws-humantask-1.1-spec_BP-16.doc    

 

 Description 

 

 

SUBMITTED BY: Martin Chapman
TITLE: Revise Specs to use RFC2119 Keywords

TARGET: BPEL4People and Human Task

DESCRIPTION: Need to revise the specifications to use RFC-2119 keywords/statements (MUST, MAY, SHOULD, MUST NOT).

PROPOSAL: For guidance look at http://docs.oasis-open.org/templates/TCHandbook/ConformanceGuidelines.html




[BP-15] What is the purpose of the optional task name param in ht:getLogicalPeopleGroup()

Created: 07/May/08  Updated: 29/Oct/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

depends on

BP-23

(Re-)Evaluation and instantiation of ...

Applied

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT, Section 6.2

DESCRIPTION: The ht:getLogicalPeopleGroup custom function is defined as
taking an optional task name as well as the name of the logical people
group. If the task name is not provided then the current task is considered.
I don't see the purpose of passing a task name here. If there is a purpose,
then we need something in the spec to make this obvious.

PROPOSAL: Remove the optional task name from the function call. This is a
simple change to the table in Section 6.2.

 

 Comments 

 

 

Comment by Dave Ings [ 10/Jun/08 07:09 AM ]

Krasimir Nedkov: Motion from Martin to defer the resolution of Issue BP-15 pending the resolution of Issue BP-6 and note the discussion we had today in JIRA. (Mark Ford is expected to propose a resolution to issue 6.)

Krasimir Nedkov: Justin seconded.

Comment by Luc Clement [ 16/Jul/08 12:20 PM ]

Proposal Assigned to Matthias Kloppmann, 20081709

Comment by Dave Ings [ 18/Sep/08 05:15 PM ]

Resolved at September 2008 F2F, see 9/18 minutes for details.




[BP-14] Need to provide access to the Outcome field

Created: 07/May/08  Updated: 11/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ralf Mueller

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Relates to

relates to

BP-24

getOutcome function missing in B4P spec

Applied

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT, Section 6.1

DESCRIPTION: The outcome field for a task is defined in Section 4.2 as a
query that is run against the output message in order to produce an
xs:string that represents the business result of the task. This field
doesn't appear in the task view described in Section 6.1 where an
application can get a list of tasks. It seems like a simple omission from
the list of fields that come back when requesting a listing of tasks.

I also think it's worth considering adding a custom function to b4p to get
the outcome from a people activity but this will be opened as a separate
issue pending the outcome of this issue (no pun intended).

PROPOSAL:

WS-HT spec:
==========

Section 6.1.1 Participant Operations:
-------------------------------------

- Add the following Operation:

Operation Name: getOutcome
Description: Get the outcome of a task
Parameters: In: task identifier, Out: string
Authorization: Any


Section 6.1.3 Advanced Query Operation:
-------------------------------------------

- Add another element to the XML-schema type "tTaskQueryResultRow"

<xsd:element name="outcome" type="xsd:string"/>

- Modify the XML-schema file "ws-humantask-api.xsd" to include above element
in complexType "tTaskQueryResultRow"

- Add a row to the 'Complete Task View' table:

Column Name: Outcome
Type: xsd:string
Constraints: -

Section 6.2 XPath Extension Functions:
----------------------------------------

Add a XPath operation:

Operation Name:
getOutcome
Description:
Returns the outcome of the task.
Evaluates to an empty string in case there is no outcome specified for the task.
If the task name is not present, the current task is considered.
Parameters:
In: task name (optional), Out: the task outcome (xsd:string)

ws-humantask-api.wsdl
================
- Add the following artifacts:

<xsd:element name="getOutcome">
<xsd:complexType>
  <xsd:sequence>
    <xsd:element name="identifier" type="xsd:anyURI" />
  </xsd:sequence>
</xsd:complexType>
</xsd:element>

<xsd:element name="getOutcomeResponse">
<xsd:complexType>
  <xsd:sequence>
    <xsd:element name="outcome" type="xsd:string" />
  </xsd:sequence>
</xsd:complexType>
</xsd:element>

<wsdl:message name="getOutcome">
<wsdl:part name="getOutcome"
                element="htdt:getOutcome" />
</wsdl:message>

<wsdl:message name="getOutcomeResponse">
<wsdl:part name="getOutcomeResponse"
                 element="htdt:getOutcomeResponse" />
</wsdl:message>


<wsdl:operation name="getOutcome">
<wsdl:input message="getOutcome"/>
<wsdl:output message="getOutcomeResponse"/>
<wsdl:fault name="illegalArgumentFault"
                 message="illegalArgumentFault"/>
</wsdl:operation>

 

 Comments 

 

 

Comment by Luc Clement [ 16/Jul/08 12:18 PM ]

Proposal Assigned to Ralf Mueller, 20081709

Comment by Luc Clement [ 23/Jul/08 10:35 AM ]

Approved as Resolved. 20080723 based on :

http://lists.oasis-open.org/archives/bpel4people/200806/msg00023.html with the following amendments:
- suggested amendment: outcomeExists => outcome
- In section 6.1.2 of WS-HT
- update outcomeExist == bool and outcome == string




[BP-13] Merge ws-humantask-context.xsd and ws-humantask-protocol.xsd into a single file

Created: 01/May/08  Updated: 28/Jun/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
 TARGET: HT, schema files

DESCRIPTION: ws-humantask-context.xsd and ws-humantask-protocol.xsd are
separate files with the same targetNamespace. They should be merged into a
single file.

PROPOSAL: Merge the content from ws-humantask-context.xsd into
ws-humantask-protocol.xsd

 

 Comments 

 

 

Comment by Dave Ings [ 11/Jun/08 04:34 AM ]

Dieter Koenig1: This is what the namespaces would look like:
 
 http://docs.oasis-open.org/ns/bpel4people/bpel4people/200803
 http://docs.oasis-open.org/ns/bpel4people/ws-humantask/200803
 http://docs.oasis-open.org/ns/bpel4people/ws-humantask/types/200803
 http://docs.oasis-open.org/ns/bpel4people/ws-humantask/api/200803
 http://docs.oasis-open.org/ns/bpel4people/ws-humantask/context/200803
 http://docs.oasis-open.org/ns/bpel4people/ws-humantask/protocol/200803
 http://docs.oasis-open.org/ns/bpel4people/ws-humantask/policy/200803

Dave Ings: Proposed resolution to issue #13: use the above namespaces.

Dave Ings: Clarification: this is a pattern, the date may change without new motions being required by the TC.

Gerhard Pfau: Dieter: Motion to resolve BP-13 as per the proposed resolution above. Martin seconds.
Accepted by unanimous consent.

Comment by Dieter Koenig [ 28/Jun/08 08:40 AM ]

Resolution incorporated into all specifications and XML artifacts, see SourceForge SVN, revision 17




[BP-12] Value of genericHumanRole in setGenericHumanRole operation call should be defined

Created: 18/Apr/08  Updated: 12/Nov/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Vinkesh Mehta

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

depends on

BP-20

Extensibility of BPEL4People and WS-H...

Applied

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT, Section 6.1.4

DESCRIPTION: The setGenericHumanRole operation in ws-humantask-api.wsdl
accepts an htd:setGenericHumanRole element. This element is defined in
ws-humantask-api-wsdl.xsd as follows:

  <xsd:element name="setGenericHumanRole">
    <xsd:complexType>
      <xsd:sequence>
        <xsd:element name="identifier" type="xsd:anyURI" />
        <xsd:element name="genericHumanRole" type="xsd:string" />
        <xsd:element name="organizationalEntity"
          type="htd:tOrganizationalEntity" />
      </xsd:sequence>
    </xsd:complexType>
  </xsd:element>

The spec doesn't define the possible values for genericHumanRole. The
genericHumanRole should either be a schema type that contains an enumerated
list of values or the spec should list what the well known role names are.
Without a formal definition in place it's going to be an interop issue.

This needs to be an open list to allow a user to extend the list of possible values

PROPOSAL:

At the 29 Oct 2008, we accepted the proposal submitted by Dieter at http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200809/msg00039.html.




[BP-11] Initialization of process generic human roles in b4p

Created: 18/Apr/08  Updated: 11/Mar/09

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ralf Mueller

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: B4P, Section 3.1 Generic Human Roles

DESCRIPTION: The b4p:peopleAssignments element is a required extension
element according to the syntax on line 292-294 in the B4P spec.

This element is defined on line 375:

<b4p:peopleAssignments>
  <htd:genericHumanRole>+
    <htd:from>...</htd:from>
  </htd:genericHumanRole>
<b4p:peopleAssignments>


With a detailed example on line 384

<b4p:peopleAssignments>

  <b4p:processInitiator>?
    <htd:from ...>...</htd:from>
  </b4p:processInitiator>

  <b4p:processStakeholders>?
    <htd:from ...>...</htd:from>
  </b4p:processStakeholders>

  <b4p:businessAdministrators>?
    <htd:from ...>...</htd:from>
  </b4p:businessAdministrators>
  
</b4p:peopleAssignments>

The initial syntax indicates that at least one role must be present. The
example immediately following it shows them as optional. My understanding is
that you have to provide at least one but it's up to you to pick which one
you want to declare in your process.

Can the initialization of these roles make use of instance data within the
process, either in the form of an expression or as arguments to a logical
people group? It seems like they wouldn't be able to since their
intialization is during the process scope's intialization which occurs prior
to the execution of the create instance activity. If there isn't any access
to instance data, then the definitions are essentially using static data. If
this is the case, then why do we require their declaration? It seems that
the infrastructure could handle the assignment of users to these roles. The
only one that doesn't fit well is the process stakeholders since that
intialization would likely require access to the process instance data.

PROPOSAL:


 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:52 AM ]

Approved as Opened 20080430

Comment by Dave Ings [ 09/Jun/08 10:11 AM ]

Extracted from 6/9 meeting minutes:

Dieter Koenig: Assigning people to process-related generic human roles happens after BPEL process initialization (see [WS-BPEL 2.0]). A BPEL4People processor MUST initialize process-related generic human roles after the end of the initial start activity of the process and before processing other activities. If that initialization fails then the fault b4p:initializationFailure MUST be thrown by a BPEL4People processor.

Dieter Koenig: (this will become the new section 3.1.2 Initialization Behavior

Gershon Janssen: Martin moves a motion: Resolve issue 11 to replacing the last paragraph of 3.1.1 with a new paragraph 3.1.2 as defined above.

Gershon Janssen: Frank seconds the motion. Accepted by unanimous consent.

Comment by Dave Ings [ 10/Jun/08 08:33 AM ]

Clarifying comment:

... and before processing other activities or links leaving the start activity.

(editors can clean this language up).




[BP-10] More control needed for the propagation of attachments

Created: 18/Apr/08  Updated: 30/Jul/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Luc Clement

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: B4P, Section 3.3 Ad-hoc Attachments

DESCRIPTION: The existing support for passing attachments from a process to
a people activity is very coarse grained. The attachment propagation element
allows for the passing of all or none of the process attachments. This makes
it impossible to control which attachments are propagated to specific people
activities within a process. Consider the use case where there are two
people activities in the same process that require different attachments.
There is no way to pass only the attachments each one needs. The workaround
is to either pass all and let the client application (or end user) filter
out the unrelated attachments or to refactor the people activities such that
they exist in separate processes. Neither solution seems good.

The existing HT protocol supports passing individual attachments to a remote
task but B4P doesn't take advantage of this.

PROPOSAL:
Update the first para of Section 3.3 changing:

"Processes can have ad-hoc attachments. It is possible to exchange ad-hoc attachments between people activities of a process, and even propagate ad-hoc attachments to and from the process level."

with the following:

"It is possible to exchange ad-hoc attachments between people activities of a process by propagating ad-hoc attachments to and from the process level."

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:52 AM ]

Approved as Opened 20080430

Comment by Dave Ings [ 10/Jun/08 05:34 PM ]

Resolve by adding text clarifying the extent and manner by which ad hoc attachments can be transferred.

The agreed to chat room text will be added as a comment once received from the scribe.

Comment by Dave Ings [ 12/Jun/08 11:32 AM ]

Dave Ings: Proposed resolution: Revised B4P with language similar to the above to clarify the current level of attachments supports.
Mark Ford: It is possible to exchange ad-hoc attachments between people activities of a process by propagating ad-hoc attachments to and from the process level.
Dave Ings: (Mark just pasted proposed text, editors to complete revision)
Krasimir Nedkov: Motion by mark to adopt the proposal for resolution of Issue BP-10. Gerhard seconded. Accepted with unanimous consent.




[BP-9] ht:getTaskID() is not defined

Created: 16/Apr/08  Updated: 02/Jul/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT, Section 6.2 Xpath Extension Functions

DESCRIPTION: There is no custom function for getting the id of a task. I
seem to recall there being function to do this in a previous version of the
spec. There's a reference made to this function in one of the samples on
line 1466. It seems like a valuable function to have and should be defined
along with the other custom functions.

PROPOSAL: Add a definition for the function to Section 6.2. Formal proposal
to follow if issue is opened.

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:52 AM ]

Approved as Opened 20080430

Comment by Dave Ings [ 10/Jun/08 09:23 AM ]

Dave Ings: Proposed resolution for 9: revise the sample (near line 1466) to remove all reference to this API call.

Dave Ings: clarification: revise the sample in WS-HT

Krasimir Nedkov: Marks moves to accept the proposed resolution. Ralf seconds. Issue carried with unanimous consent.

Comment by Ivana Trickovic [ 25/Jun/08 09:36 AM ]

Resolution incorporated into the specification




[BP-8] Are string literals required for part, task, and LPG name in HT functions?

Created: 16/Apr/08  Updated: 02/Jul/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT, Section 6.2 Xpath Extension Functions

DESCRIPTION: It's not clear whether the functions require string literals
for the part, task, or LPG name params or whether callers can use
expressions for these values. In the ws-bpel 2.0 spec there are similar
instances where functions that accept the name of a resource (i.e.
bpel:getVariableProperty ) require that the name of the resource is passed
as a string literal. This is beneficial for static analysis and as well as
makes the behavior of the process/task easier to follow.

PROPOSAL: Add some text to the description of the Xpath extensions to
indicate that the part, task, and LPG params must be passed as string
literals. A similar change is required to the B4P spec. Formal proposal to
follow if the issue is opened.

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:52 AM ]

Approved as Opened 20080430

Comment by Dave Ings [ 10/Jun/08 06:53 AM ]

Dave Ings: To be consistent with BPEL 2.0, in section WS-HT 6.2, add the following introductory text: Parameters that specify part, task, or LPG names passed to WS-HumanTask XPath extension functions must be literal strings. All other parameters don't have that restriction. For B4P, in section 5.0, add the following introductory text: Parameters that specify people activity names passed to BPEL4People XPath extension functions must be literal strings. All other parameters don't have that restriction.

Dave Ings: (The editors may clean this text up.)

Comment by Ivana Trickovic [ 25/Jun/08 09:36 AM ]

Resolution incorporated into the specification




[BP-7] addAttachment API call is missing the content type param

Created: 16/Apr/08  Updated: 16/Jul/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Vinkesh Mehta

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT API

DESCRIPTION: The element defined for the addAttachment operation does not
have a means for indicating the content type.

PROPOSAL: Add the attachment type to the element. This involves a change to
ws-humantask-api.wsdl

Current definition:

  <xsd:element name="addAttachment">
    <xsd:complexType>
      <xsd:sequence>
        <xsd:element name="identifier" type="xsd:anyURI" />
        <xsd:element name="name" type="xsd:string" />
        <xsd:element name="accessType" type="xsd:string" />
        <xsd:element name="attachment" type="xsd:anyType" />
      </xsd:sequence>
    </xsd:complexType>
  </xsd:element>

Proposed definition:

  <xsd:element name="addAttachment">
    <xsd:complexType>
      <xsd:sequence>
        <xsd:element name="identifier" type="xsd:anyURI" />
        <xsd:element name="name" type="xsd:string" />
        <xsd:element name="accessType" type="xsd:string" />
        <xsd:element name="contentType" type="xsd:string" />
        <xsd:element name="attachment" type="xsd:anyType" />
      </xsd:sequence>
    </xsd:complexType>
  </xsd:element>

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:51 AM ]

Approved as Opened and Resolved 20080430




[BP-6] Does ht:getLogicalPeopleGroup() cause an LPG to evaluate?

Created: 15/Apr/08  Updated: 29/Oct/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Dieter Koenig

Resolution:

Unresolved

Votes:

0

 

Issue Links:

Depends on

depends on

BP-23

(Re-)Evaluation and instantiation of ...

Applied

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT and B4P, Section 3.2.1 Using Logical People Groups

DESCRIPTION: Should the use of the custom function
ht:getLogicalPeopleGroup() or b4p:getLogicalPeopleGroup() cause the logical
people group to evaluate or does it return the value from a previous
evaluation? My initial take on these functions was that the use of the
function DOES NOT trigger an evaluation of the logical people group. This is
based on the fact that there is no way to pass arguments to the LPG for its
evaluation. There is also some text in the b4p spec which could be read to
support this. For example:

--- begin quote Lines 424-426 from Section 3.2.1 of B4P: ---
If the result of a previous people query needs to be re-used, then this
result needs to be referenced explicitly from the process context. Please
refer to section 5 "XPath Extension Functions" for a description of the
syntax.
--- end quote ---

If the intent of the function is to NOT trigger an evaluation, then this
should be made explicit in the spec. If the intent is to have this function
trigger an evaluation, then this should be made explicit and the function
should support arguments.

PROPOSAL: Make it clear in the spec that the use of this function causes the
LPG to evaluate in the same way that the ht:from and bpel:from work. Add
support for optional named arguments on the functions. Formal proposal to
follow if issue is opened.

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:52 AM ]

Approved as Opened 20080430

Comment by Luc Clement [ 16/Jul/08 12:20 PM ]

Proposal Assigned to Matthias Kloppmann, 20081709

Comment by Dave Ings [ 18/Sep/08 05:14 PM ]

Resolved at September 2008 F2F, see 9/18 minutes for details.




[BP-5] Discrepancy between spec and schema regarding LPG attribute

Created: 11/Apr/08  Updated: 02/Jul/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT and B4P, Section 3.2.1 Using Logical People Groups

DESCRIPTION: Both the ht:from and bpel:from have a logicalPeopleGroup
attribute (b4p:logicalPeopleGroup in the latter). In the specification
examples and pseudo schema this attribute is an NCName but the schema files
have this attribute as a Qname. It seems like the value should be a Qname.
If it's an NCName then there is the possibility of conflicts if two imported
humanInteraction documents define the same logicalPeopleGroup name.

PROPOSAL: Change the examples in the spec from NCName to Qname for
logicalPeopleGroup attribute. Exact line numbers to be provided as part of
the formal proposal.

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:52 AM ]

Approved as Opened 20080430

Comment by Dave Ings [ 10/Jun/08 06:35 AM ]

Dave Ings: Proposed resolution: To be consistent with the resolution of issue 4, the schema needs to be changed to logical people attribute of type NCName. In ws-humantask.xsd it is line 465.

Krasimir Nedkov: Motion from Gerhard to accept the proposed resolution. Ralf seconds. Accepter by unanimous concent.

Comment by Ivana Trickovic [ 25/Jun/08 09:35 AM ]

Resolution incorporated into the specification




[BP-4] Declaration of imported LPG not mentioned in B4P spec

Created: 11/Apr/08  Updated: 02/Jul/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Ivana Trickovic

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: HT, Section 3.2.1 Using Logical People Groups

DESCRIPTION: Constellation 3 provides the ability to import a human
interaction into a bpel process. In this scenario, the imported human
interaction may contain one or more logicalPeopleGroup definitions. Does the
import of a human interaction constitute an implicit declaration of a
logicalPeopleGroup within the bpel process or is an explicit declaration
required? Some type of declaration is necessary since these LPG's are
stateful. Consider the following example consisting of a standalone human
interaction file imported into a bpel process.

<htd:humanInteractions targetNamespace="urn:example" ...>
   ...
   <htd:logicalPeopleGroups>
      <htd:logicalPeopleGroup name="LPG-1">
         ...
      </htd:logicalPeopleGroup>
   </htd:logicalPeopleGroups>
   <htd:tasks>
      <htd:task name="T-1">
         ...
         <htd:potentialOwners>
            <htd:from logicalPeopleGroup="LPG-1">
               ...
            </htd:from>
         </htd:potentialOwners>
         ...
      </htd:task>
      ...
    </htd:tasks>
</htd:humanInteractions>

<bpel:process xmlns:ex="urn:example" ...>
   ...
   <bpel:import location="example.htd" namespace="urn:example" .../>
   ...

   <b4p:humanInteractions>
      <htd:logicalPeopleGroups>
         <htd:logicalPeopleGroup name="LPG-1" reference="ex:LPG-1"/>
      </htd:logicalPeopleGroups>
   </b4p:humanInteractions>

   ...

   <b4p:peopleActivity ...>
      <b4p:localTask reference="ex:T-1"/>
   </b4p:peopleActivity>

   ...

</bpel:process>

The declaration seems redundant but in the case where the process is only
utilizing a subset of the imported logicalPeopleGroups it better conveys to
the deployer which logicalPeopleGroups need to be bound to people queries.

PROPOSAL: none

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:52 AM ]

Approved as Opened 20080430

Comment by Dave Ings [ 10/Jun/08 04:57 AM ]

Dave Ings: Motion to revise 2.4.2, bullet "logical people group" with the above sentence, to resolve issue 4.

Dave Ings: The reference attribute MUST specify a logical people group, in case that a logical people group is used that is defined in an imported WS-HumanTask definition. The list of logical people groups MUST contain those logical people groups that are used in the human interaction definition.

Frank Leymann: Gerhard tables the corresponding motion, Alessandro seconds.

Dave Ings: Correction: Dieter was the seconded.

Comment by Dave Ings [ 10/Jun/08 08:16 AM ]

The solution refers to 2.4.2 in WS-HumanTask.

Comment by Dave Ings [ 10/Jun/08 08:20 AM ]

Clarifying comment re the list of logical people groups: you would only have to declare those LPGs that were being referenced by the human task that the process is using.

Comment by Ivana Trickovic [ 25/Jun/08 09:35 AM ]

Resolution incorporated into the specification




[BP-3] Syntax for LPG in B4P spec missing reference attribute

Created: 08/Apr/08  Updated: 18/Jun/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Vinkesh Mehta

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: B4P, Section 2.4.1 Syntax

Change the existing LPG sample from:

    <htd:logicalPeopleGroups/>?
      <htd:logicalPeopleGroup name="NCName">+
        ...
      </htd:logicalPeopleGroup>
    </htd:logicalPeopleGroups>

To:

    <htd:logicalPeopleGroups/>?
      <htd:logicalPeopleGroup name="NCName" reference="QName"?>+
        ...
      </htd:logicalPeopleGroup>
    </htd:logicalPeopleGroups>

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:50 AM ]

Approved as Opened and Resoved 20080430




[BP-2] Defer Activation Time is missing from HT protocol message

Created: 02/Apr/08  Updated: 29/Oct/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Michael Rowley

Resolution:

Incomplete

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: WS-HT
 
DESCRIPTION: There is no mechanism for passing the defer activation time to a remote task. This value should be part of the humanTaskContext that is defined in Section 7.4.1 of the HT specification. It is important to include this value since the task needs to be able to enter the task management system in a CREATED state such that an admin could select the task and activate it prior to the defer activation time.
 
PROPOSAL: Update the humanTaskContext to include the defer activation value. This involves the following change to ws-humantask-protocol.xsd:
 
Existing definition -
 
  <xsd:complexType name="tHumanTaskContext">
    <xsd:sequence>
      <xsd:element name="priority" type="xsd:nonNegativeInteger"
        minOccurs="0" />
      <xsd:element name="peopleAssignments" type="tPeopleAssignments"
        minOccurs="0" />
      <xsd:element name="isSkipable" type="xsd:boolean" minOccurs="0" />
      <xsd:element name="expirationTime" type="xsd:dateTime"
        minOccurs="0" />
      <xsd:element name="attachments" type="tAttachments"
        minOccurs="0" />
    </xsd:sequence>
  </xsd:complexType>
 
Proposed definition -
 
  <xsd:complexType name="tHumanTaskContext">
    <xsd:sequence>
      <xsd:element name="priority" type="xsd:nonNegativeInteger"
        minOccurs="0" />
      <xsd:element name="peopleAssignments" type="tPeopleAssignments"
        minOccurs="0" />
      <xsd:element name="isSkipable" type="xsd:boolean" minOccurs="0" />
      <xsd:element name="deferActivationTime" type="xsd:dateTime"
        minOccurs="0" />
      <xsd:element name="expirationTime" type="xsd:dateTime"
        minOccurs="0" />
      <xsd:element name="attachments" type="tAttachments"
        minOccurs="0" />
    </xsd:sequence>
  </xsd:complexType>

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:49 AM ]

Approved as Opened 20080430

Comment by Luc Clement [ 14/May/08 11:12 AM ]

This was resolved 20080515 as posted by Mark Ford at http://www.oasis-open.org/apps/org/workgroup/bpel4people/email/archives/200805/msg00010.html

Comment by Dieter Koenig [ 28/Jun/08 08:39 AM ]

Resolution incorporated into the WS-HumanTask specification and the ws-humantask-context.xsd, see SourgeForge SVN, revision 18

Comment by Luc Clement [ 09/Jul/08 01:15 PM ]

Per http://lists.oasis-open.org/archives/bpel4people/200807/msg00000.html we need to reopen BP-2.

During the Editor's review we uncovered an issue with the definition of deferActivation in ws-ht. The use of the DeferActivationTime in ws-ht is currently specified as xsd:dateTime but the b4p spec specifies b4p:for and b4p:until elements to specify a duration and a point in time respectively. The ws-ht spec does not account for this.

During the 20080709 TC call, we agreed to reopen BP-2

Comment by Luc Clement [ 20/Aug/08 10:32 AM ]

During the 20080820 TC Call we resolve this item as follow: "Amendment to BP-2: We want to say explicitly in B4P that the people activity is responsible for computing an absolute time from the duration, rather than leaving it up to the task."




[BP-1] Remove fault name from getFault HT API call

Created: 02/Apr/08  Updated: 16/Jul/08

Status:

Applied

Project:

OASIS BPEL4People TC

Component/s:

None

Affects Version/s:

None

Fix Version/s:

None

 

Type:

Improvement

Priority:

Minor

Reporter:

Luc Clement

Assigned To:

Vinkesh Mehta

Resolution:

Unresolved

Votes:

0

 

 

 Description 

 

 

SUBMITTED BY: Mark Ford
TARGET: WS-HT API

DESCRIPTION: The element "getFault" used in the getFault operation defined
in the HT API has a required faultName child element. A task only supports
having a single fault set on it so it doesn't make sense to accept a fault
name here. This is likely left over from a previous version of this schema
where faultName was removed from the call but left behind in the Schema. The
description for getFault in the table of operations in Section 6 does not
have the fault name as an input parameter.

PROPOSAL: Remove the faultName child element from the getFault element in
ws-humantask-api.wsdl as follows:

Existing definition -

  <xsd:element name="getFault">
    <xsd:complexType>
      <xsd:sequence>
        <xsd:element name="identifier" type="xsd:anyURI" />
        <xsd:element name="faultName" type="xsd:NCName" />
      </xsd:sequence>
    </xsd:complexType>
  </xsd:element>

New definition -

  <xsd:element name="getFault">
    <xsd:complexType>
      <xsd:sequence>
        <xsd:element name="identifier" type="xsd:anyURI" />
      </xsd:sequence>
    </xsd:complexType>
  </xsd:element>

 

 Comments 

 

 

Comment by Luc Clement [ 30/Apr/08 11:32 AM ]

Moved to Resolved state: 20080430

Comment by Luc Clement [ 16/Jul/08 11:21 AM ]

The change was applied to ws-humantask-api.wsdl (and not from xsd as suggested in the proposal)





Generated at Tue Feb 07 02:20:49 CST 2012 using JIRA 187.