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

 


Help: OASIS Mailing Lists Help | MarkMail Help

ws-calendar message

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


Subject: [OASIS Issue Tracker] Commented: (WSCALENDAR-308) PR02 What are youdefining? XCAL? WS-Calendar? Some combination?



    [ http://tools.oasis-open.org/issues/browse/WSCALENDAR-308?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=25526#action_25526 ] 

Toby Considine commented on WSCALENDAR-308:
-------------------------------------------

This is the tip of a large unresolved issue that redounds throughout the document.

XCAL is defined elsewhere in the IETF.Ws-Calendar schemas are not XCAL, though they should product artifacts that are compatible with XCAL. What does "compatible" mean? Does it mean identical, representative, conformable, or transformable?

In XCAL, just about every item is a string expressed as an "xcal:text". This is the enemy of automated processing. In *almost* every case, a xcal:text which conforms to the xcal parsing rules is a valid xml artifact of the nominally identical format, e.g. xcal:durationType and xs:duration. In a few cases, there is no equivalence--xcal:bool and xs:boolean do not have equivalent string representations.

In other places we hav skated around confronting the issues. Ws-Calendar payloads started out as placing an "anystring" inside attach. They then became the ugly named x-ws-calendar-attach as a different parameter. Perhaps it would have been better to define Payload as an extension to Attach. This might the expressions of payloads with a collection of intervals more approachable for implementers.

Temporal Relations do not appear in the schemas right now. We perhaps should formally extend the enumeration of RelType to include the temporal relations.

Vavailability suffers from the current approach. Long-standing and well known vcomponents exist as a bag of properties bag of parameters, and a bag of components. This creates great flexibility and allows compatibility with an immense installed base. Vavailability has no installed base. There is no compatibility reason why we cannot define it strictly here. And yet...

Inside Energy Interop, we have discovered that we need the exceptions for vavailability. "I am available 2-3 MWF and from 3:00 to 6:00 on Thursdays stating on Memorial Day and running through the end of the Summer" is concisely expressed in vavailability. "I am also available all day on on June 15th" is not so easily expresable—nor is "but unavailable on July 4 and on Bastille Day". Perhaps the same trick of recursion that is used to allow any component in anything would answer the knock-outs that modify the recurrence expressed in vavailability.


> PR02 What are you defining? XCAL? WS-Calendar? Some combination?
> ----------------------------------------------------------------
>
>                 Key: WSCALENDAR-308
>                 URL: http://tools.oasis-open.org/issues/browse/WSCALENDAR-308
>             Project: OASIS Web Services Calendar (WS-Calendar) TC
>          Issue Type: Sub-task
>          Components: spec
>    Affects Versions: pr02
>         Environment: William Cox
>            Reporter: William Cox
>            Assignee: Toby Considine
>             Fix For: wd20
>
>
> The document is confusing in several places as to what is being defined. The schemas define all of xcal, the document talks mostly of WS-Calendar (which is now structured as an extension to xcal).
> Line 368 demonstrates this with the "Use in WS-Calendar" contasting XCAL and WS-Calendar. Likewise Line 657 "Text" "Discussion" says "The use of text in WS-Calendar is not defined"
> So which is being specified? This relates to another issue, WSCALENDAR-305, which recommends a simple model (Platform Independent Model) that is compatible with the broader model.
> Table headings don't help. Table 3-1 headings are "element, use, use in WS-Calendar"
> But is the broader model xcal or ws-calendar? Or is ws-calendar a profile on xcal?
> This is a general description and discussion issue throughout the document.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://tools.oasis-open.org/issues/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira




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