13:47:23 <RRSAgent> RRSAgent has joined #forms
13:47:24 <RRSAgent> logging to http://www.w3.org/2017/02/15-forms-irc
13:47:25 <trackbot> RRSAgent, make logs public
13:47:26 <Zakim> Zakim has joined #forms
13:47:28 <trackbot> Zakim, this will be
13:47:28 <Zakim> I don't understand 'this will be', trackbot
13:47:29 <trackbot> Meeting: XForms Users Community Group Teleconference
13:47:29 <trackbot> Date: 15 February 2017
13:47:48 <Steven> Agenda: https://lists.w3.org/Archives/Public/public-xformsusers/2017Feb/0019
13:47:57 <Steven> Steven has changed the topic to: Agenda: https://lists.w3.org/Archives/Public/public-xformsusers/2017Feb/0019
13:48:04 <Steven> Chair: Steven
13:48:13 <Steven> Regrets: Alain, Philip
13:55:42 <ebruchez> ebruchez has joined #forms
14:04:18 <Steven> Topic: JSON content
14:04:18 <Steven> http://lists.w3.org/Archives/Public/public-xformsusers/2017Feb/0004.html
14:04:44 <Steven> Steven: I did the edits, fairly striaghtforward, and I left it vague about what it means for a JSON value to enter this system.
14:05:15 <Steven>  ACTION-2111?
14:05:15 <trackbot> ACTION-2111 -- Steven Pemberton to Refine the spectext about recognising if a incoming alue is json or not. -- due 2017-02-15 -- OPEN
14:05:15 <trackbot> https://www.w3.org/2005/06/tracker/xforms/actions/2111
14:05:20 <Steven> close  ACTION-2111
14:05:20 <trackbot> Closed ACTION-2111.
14:05:32 <Steven> Topic: Rebuild
14:05:32 <Steven> https://lists.w3.org/Archives/Public/public-xformsusers/2017Feb/0012.html
14:06:03 <Steven> * At the start of a recalculate, there is a, possibly empty, list L of
14:06:03 <Steven> instance nodes whose values may have been changed, e.g., by user input or
14:06:03 <Steven> by a setvalue action.
14:06:03 <Steven> * All nodes in L, if any, and nodes that are computationally dependent on
14:06:04 <Steven> nodes in L must be recalculated.
14:06:04 <Steven> * Only a single recalculation of each compute that is computationally
14:06:04 <Steven> dependent on elements in L may be performed.
14:06:05 <Steven> * A compute must be recalculated before any computes that are
14:06:05 <Steven> computationally dependent on the instance node associated with it.
14:06:05 <Steven> * If a compute is computationally dependent on an element in L and part of
14:06:05 <Steven> a circular dependency, then an xforms-compute-error event must be
14:06:05 <Steven> dispatched to the model element.
14:06:41 <Steven> Steven: I suspect we can use the same sort of language with rebuild as well
14:07:13 <Steven> Erik: Yeah
14:09:29 <Steven> Steven: Then that would allow different implementations, just as you Erik have combined recalc with revalidate.
14:10:56 <Steven> Erik: We have to say something about how the graph can depend on single values; we shouldn;t make life hard for implementors
14:11:47 <Steven> Steven: I agree; but as long as we specify it right, implementors can implement just as they like. However, we should warn for the edgecases
14:12:29 <Steven> Steven: It's the section of the spec I'm working on, so I'll give it a try.
14:12:39 <Steven> Topic: Serialization as multipart/related
14:12:39 <Steven> https://lists.w3.org/Archives/Public/public-xformsusers/2016Dec/0006.html
14:12:39 <Steven> https://lists.w3.org/Archives/Public/public-xformsusers/2016Dec/0030.html
14:14:19 <Steven> Steven: So one possiblity could be having a 'magic' data type, that holds the string value of a URI, but on submit it is recognised and dereferenced and the binary submitted.
14:16:17 <Steven> Erik: That solves one of the problems, but not all of them.
14:17:35 <Steven> Erik: I was wondering if there are other 'magic' things in the spec.
14:19:37 <ebruchez> "Subsequent part requirements"
14:19:42 <ebruchez> 11.5.7.2 Serialization as multipart/related
14:21:56 <Steven> Erik: THis section doesn't say where some of the values come from.
14:22:03 <Steven> s/TH/Th/
14:23:11 <Steven> ... it could be stored in the data, or out of band.
14:24:50 <Steven> Steven: Do you have any recommendations?
14:25:44 <Steven> Erik: So we need information about the to-be-dereferenced URI, and then details about the parts
14:25:52 <Steven> ... we could say it is implementation dependent
14:26:06 <Steven> s/dep/-dep/
14:26:31 <Steven> Steven: That would make it hard to write interoperable forms
14:27:00 <Steven> Erik: We could clarify that the purpose is to handle things populated by upload
14:28:54 <Steven> ... a piece of text that says the type of the attachment is stored at the URI is populated in the instance.
14:30:44 <Steven> Steven: I need longer to think about this.
14:31:24 <Steven> ... if you could send an email with details of how you might implement it, that might help to crystallise the thoughts.
14:31:51 <Steven> Erik: We do one of the muliti-parts, and we figure out how to identify the parts. I'll send an email.
14:32:15 <Steven> Topic: AOB
14:32:31 <Steven> Erik: You went to XML Prague.
14:35:09 <Steven> Steven: Well attended. The proceedings are online.
14:36:55 <Steven> ... XPath 3 is just about ready.
14:38:12 <Steven> [ADJOURN]
14:40:03 <Steven> rrsagent, make minutes
14:40:03 <RRSAgent> I have made the request to generate http://www.w3.org/2017/02/15-forms-minutes.html Steven
14:42:18 <Steven> Present: Erik, Steven
14:42:19 <Steven> rrsagent, make minutes
14:42:19 <RRSAgent> I have made the request to generate http://www.w3.org/2017/02/15-forms-minutes.html Steven
16:58:52 <ebruchez> ebruchez has joined #forms
17:11:36 <Zakim> Zakim has left #forms