12:39:53 <RRSAgent> RRSAgent has joined #forms
12:39:53 <RRSAgent> logging to http://www.w3.org/2016/10/12-forms-irc
12:39:55 <trackbot> RRSAgent, make logs public
12:39:55 <Zakim> Zakim has joined #forms
12:39:57 <trackbot> Zakim, this will be
12:39:57 <Zakim> I don't understand 'this will be', trackbot
12:39:58 <trackbot> Meeting: XForms Users Community Group Teleconference
12:39:58 <trackbot> Date: 12 October 2016
12:40:20 <Steven> Agenda: https://lists.w3.org/Archives/Public/public-xformsusers/2016Oct/0011
12:40:32 <Steven> Steven has changed the topic to: Agenda: https://lists.w3.org/Archives/Public/public-xformsusers/2016Oct/0011
12:40:44 <Steven> Chair: Steven
12:42:33 <alain> alain has joined #forms
12:58:46 <ebruchez> ebruchez has joined #forms
13:03:21 <Steven> Present: Steven, Alain, Erik
13:05:50 <Steven> https://www.kadaster.nl/web/english/About-Kadaster.htm
13:07:12 <Steven> Topic: Advance warning, call with be at 14:00 in Europe on 2 Nov
13:07:12 <Steven> http://www.timeanddate.com/worldclock/fixedtime.html?msg=XForms+Call&iso=20161102T09&p1=43&ah=1
13:07:29 <Steven> Steven: Only one week this year.
13:07:43 <Steven> ... one hour earlier in Europe
13:07:56 <Steven> Topic: Section 9
13:07:56 <Steven> https://lists.w3.org/Archives/Public/public-xformsusers/2016Oct/0009.html
13:08:10 <Steven> Steven: I'm assuming Section 9 is now done.
13:08:20 <Steven> Topic: @value
13:08:20 <Steven> https://lists.w3.org/Archives/Public/public-xformsusers/2016Oct/0010.html
13:09:50 <Steven> Steven: Surprising amount of work. Many differences are just difference in text, which I can put right. But I still have questions.
13:09:53 <Steven> ... Deprecate submission/header/name, submission/header/value?
13:11:08 <Steven> ... I realise now that you can generate several headers with one name using several values.
13:11:25 <Steven> ... So that is reason enough to keep header/value
13:12:08 <Steven> Erik: Header/name /value look like item/name /value
13:12:12 <pfennell> pfennell has joined #forms
13:12:27 <Steven> ... I have no strong feelings about deprecating
13:12:36 <Steven> Steven: Fine. I thought it was worth mentioning.
13:12:56 <Steven> Steven: What happens if the evaluation of @value that has type ANY fails?
13:13:06 <Steven> Present+Philip
13:14:41 <Steven> Erik: Hmm, good question
13:15:06 <Steven> ... for bindings we fall back, and dispatch an error
13:15:19 <Steven> ... for actions we have xforms-action-error
13:15:57 <Steven> ... we don't want actions to silently fail
13:16:41 <Steven> ... one solution would be to dispatch an event
13:16:53 <Steven> ... maybe we already do this already at a general level
13:17:11 <ebruchez> 10.5.2 Error Recovery Inside Actions
13:17:47 <Steven> "Any error taking place during action processing stops the outermost action handler. Errors include:
13:17:47 <Steven>  expression errors
13:17:47 <Steven>  binding errors with the bind or model attribute
13:17:47 <Steven>  binding errors with complex or readonly content
13:17:48 <Steven>  missing mandatory attributes or unsupported attribute values on action elements
13:17:48 <Steven> The xforms-action-error event is dispatched to the observer of the action. Observers include: controls and the model, instance, and submission elements.
13:17:48 <Steven> Note:
13:17:48 <Steven> Some actions silently ignore error conditions and are not subject to the processing above, including:
13:17:49 <Steven> setvalue pointing to the empty sequence or to an atomic item (such as a string) instead of a node
13:17:51 <Steven> delete with the empty sequence or an empty overridden context
13:17:52 <Steven> insert with an empty or non-element insert context, an empty overridden context, or an empty origin
13:17:54 <Steven>  actions with AVTs evaluating to the empty sequence
13:17:55 <Steven> dispatch, send, setfocus, setindex, and toggle when the target element is not found"
13:17:57 <Steven> https://www.w3.org/community/xformsusers/wiki/index.php?title=XForms_2.0#Error_Recovery_Inside_Actions
13:19:39 <Steven> Steven: So that covers action/param, and dispatch/property. What about function/param
13:19:42 <Steven> ?
13:19:56 <Steven> s/param/result
13:20:42 <Steven> Erik: I propose that we use the same processing as bind errors
13:23:09 <Steven> s/I propose that we use the same processing as bind errors//
13:24:15 <Steven> Erik: Right now it is implementation dependent for when the function elements are processed.
13:26:01 <Steven> ... there can be preprocessing, that spots errors, or a dynamic error
13:30:04 <Steven> Steven: There are lots of places where preprocessing could occur already, and we don't say what to do in those cases.
13:36:40 <Steven> Steven: We say that a string @value returns the empty string if evaluation fails
13:37:07 <Steven> ... what should we say for other types?
13:37:15 <Steven> Erik: There is no good default value
13:37:35 <Steven> Steven: Then just say "it is undefined"?
13:45:33 <Steven> ACTION: Erik to propose text for what value is used when the evaluation of function/result@value fails
13:45:33 <trackbot> Created ACTION-2076 - Propose text for what value is used when the evaluation of function/result@value fails [on Erik Bruchez - due 2016-10-19].
13:46:25 <Steven> Steven: I think we can make @value a Common attribute
13:46:32 <Steven> ... along these lines:
13:46:33 <Steven> Generalise more?:
13:46:33 <Steven> * Returns a string or any type as specified
13:46:33 <Steven> * Evaluated in in-scope context, unless there is a binding, and then it
13:46:33 <Steven> uses that context.
13:46:34 <Steven> * If the element can have PCDATA content, then @value overrides it
13:46:34 <Steven> * If evaluation fails, for strings the empty string is used; for ANY ???
13:46:34 <Steven> (decision)
13:47:35 <Steven> Steven: Then I can put the description of @value in one central place, and all the places where it is used, they use that Common attribute
13:47:48 <Steven> Erik: Finer to do that.
13:47:54 <Steven> s/Finer/Fine/
13:50:00 <Steven> Erik: I would prefer it if section 10.5.2 supercedes "returning the empty string".
13:50:06 <Steven> Steven: That's fine.
13:51:18 <Steven> Topic: AOB
13:51:22 <Steven> [None]
13:51:39 <Steven> [ADJOURN]
13:51:44 <Steven> rrsagent, make minutes
13:51:44 <RRSAgent> I have made the request to generate http://www.w3.org/2016/10/12-forms-minutes.html Steven
15:17:45 <Zakim> Zakim has left #forms
15:55:29 <ebruchez> ebruchez has joined #forms