OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

xdi message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: Minutes: XDI TC Telecon Friday 2012-08-03


Following are the minutes of the unofficial telecon of the XDI TC at:

Date:  Friday, 03 August 2012 USA
Time:  9:00AM - 10:30AM Pacific Time (16:00-17:30 UTC)

ATTENDING

Les Chasen
Markus Sabadello
Drummond Reed
Bill Barnhill
Giovanni Bartolomeo
Phil Windley


THE ETHERPAD LINK FOR TODAY IS:
     http://piratepad.ca/PDyUPGEBbN


AGENDA


***** NEWS & UPDATES *****

--- ODATA UPDATE

Giovanni reported that he attended the last OData meeting remotely and is willing to help with collaboration.


--- WIKI HOME PAGE

Markus and Drummond have completed the cleanup and refactoring of the home page and all the Category pages. During the call we also added an ActionItems page to track major action items:

   https://wiki.oasis-open.org/xdi/ActionItems


***** DECISION POINT QUEUE REVIEW *****

The decision queue stack is now shown on the following three auto-generated Category pages:

   https://wiki.oasis-open.org/xdi/CategoryLastCall
   https://wiki.oasis-open.org/xdi/CategoryCurrent
   https://wiki.oasis-open.org/xdi/CategoryHighPriority

Since we did not close on any decisions on this call, we agreed to focus next week's call on discussions about multiplicity and the API $add operations.


***** DECISION POINTS FOR THIS CALL *****

--- MULTIPLICITY (DRUMMOND)

   https://wiki.oasis-open.org/xdi/XdiMultiplicity

We walked through the proposal. Bill's key concern is that it does not allow every context node to be treated as a set in the mathematical sense. Drummond's POV is that it is a feature to be able to definitely declare which context nodes (sets) are singletons.

We looked at the "Combining Attribute Singletons and Collections" pattern on pg 14 of the current XDI Graph Patterns document (https://www.oasis-open.org/committees/download.php/46426/xdi-graph-patterns-2012-07-06.pdf) to see the benefit of being able to have both singletons and collections of the same attribute side-by-side.

# BILL to develop an alternate proposal for how to address all the use cases illustrated in the "Combining Attribute Singletons and Collections" pattern.


--- API PROPOSALS (MARKUS)

   https://wiki.oasis-open.org/xdi/CategoryApi

We reviewed the granular format in which Markus has been posting proposals to clarify key details of the API, starting with $add operations.

We had a long discussion about the format of the object of a $add operation. Three options were discussed:
   1) An XDI subject
   2) An XDI statement
   3) A reference to an XDI graph

Bill gave the analogy that a $add operation is essentially merging two XDI graphs -- the target graph identified in the message, and the source graph to be added. There was a consensus on this conceptual view. The question then comes down to the form of the source graph. The JSON serialization proposal (https://wiki.oasis-open.org/xdi/JSONSerializationRules - which is in Last Call) calls for the source graph to be included inline as a JSON object of the $add predicate.

Bill suggested and provided an example (now posted on the Discussion page -- https://wiki.oasis-open.org/xdi/JSONSerializationRules/Discussion) of how the source graph could be included by reference rather than by value. In this case the reference was to another graph included in the same message (although the syntax of such an inclusion was not clear).

Drummond felt that while the inclusion-by-reference option was a useful one, particularly for XDI graphs that are external to the message, that use case is complex and would be better served by an extension to the $add operation that clearly specified the semantics of add-by-reference, such as $add$get.

# BILL to generate a proposal for how a $add-by-reference would work.


--- TEST DATA FORMAT (BILL)

  https://wiki.oasis-open.org/xdi/TestDataFormatProposal

This is a new proposal that Bill added and which we will discuss more as soon as we are through the multiplicity and $add discussions.


***** NEXT CALL *****

To accommodate the summer holidays, we will hold the call next week at the regular time, then take 2 weeks off. So the August call schedule will be:


[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]