See also: IRC log
https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0020
https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0028.html
Steven: Did you see Philip's question?
Erik: Let me give an example (if I can think of one)
<ebruchez> <visa>some text<number/>M</visa>
Erik: <visa> is non
relevant
... but it is complex
... you can't bind to the text of visa, so should it also be
emptied?
... (because there are no empty text nodes, so the control
would become nonrelevant henceforth)
Steven: Do you have a use-case why it *shouldn't* be emptied?
Erik: I don't think so.
... Though there might be whitespace between elements, which
would get lost
<pfennell> If visa of <visa><number>123456</number></visa> is made non-relevant by an action will number lose it's text node? If number is of type xs:integer then that empty node would raise a validation error.
Philip: If the number got emptied it wouldn;t validate on the service side
Steven: That would be your own
responsibility.
... emptying the fields is an acceptable usecase; I've needed
it myself in the past.
... I proposed nonrelevant="keep|remove|empty"
Alain: Fine
Philip: OK
Erik: OK
<scribe> ACTION: Steven to add @nonrelevant according to https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0021.html [recorded in http://www.w3.org/2017/03/29-forms-minutes.html#action01]
<trackbot> Created ACTION-2115 - Add @nonrelevant according to https://lists.w3.org/archives/public/public-xformsusers/2017mar/0021.html [on Steven Pemberton - due 2017-04-05].
https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0022
Steven: Point one, does it need to be done with an event?
Erik: I've never liked this.
Steven: I'll investigate getting
rid of those events during processing.
... Point two,
https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0023.html
... His point seems to be that these details don't need to be
exposed in the spec
Erik: If you do two setvalues
after each other, there is no recalculation between the
two
... Similarly to the UI, but that may be less of a
problem.
... The alternative would be to have all actions cause an
update
Steven: Isn't it the case that
this only arises in an action?
... if we just said that the model doesn't get updated until
the end of an action, we wouldn't need to describe it with the
deferred update mechanism.
Erik: I wonder if Michael's question is just about that.
Steven: It seems if we could
express it without saying how, only what you want to happen, it
would be more along our lines of working anyway
... I'll look at it this week, and see what can be done to
simplify it.
https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0024
Steven: I don't know how much multipart/ is used
Alain: Nor do I
Erik: Would this be possible in a web browser?
Alain: I think so.
Steven: Shal we give it a try?
s/shal/shall/
Steven: or is it not important enough?
Philip: Give it a try.
Erik: I think people try to avoid
it
... but I think it is implemented on servers
... It is a complication
... but it is how browsers might send multiple uploads
... I have no good feel, but I would like to have something
like this.
... It is still a 'thing'.
<scribe> ACTION: Steven to spec up <part/> proposal. [recorded in http://www.w3.org/2017/03/29-forms-minutes.html#action02]
<trackbot> Created ACTION-2116 - Spec up <part/> proposal. [on Steven Pemberton - due 2017-04-05].
Steven: AOB
[None]
[ADJOURN]
<scribe> ACTION: Steven to investigate removing the need for deferred update flags [recorded in http://www.w3.org/2017/03/29-forms-minutes.html#action03]
<trackbot> Created ACTION-2117 - Investigate removing the need for deferred update flags [on Steven Pemberton - due 2017-04-05].
This is scribe.perl Revision: 1.152 of Date: 2017/02/06 11:04:15 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Succeeded: s/tion/tion?/ Succeeded: s/;/'/ Succeeded: s/set value/setvalue/ Succeeded: s/adte/date/ FAILED: s/shal/shall/ Present: Alain Erik Philip Steven No ScribeNick specified. Guessing ScribeNick: Steven Inferring Scribes: Steven Agenda: https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0025 Found Date: 29 Mar 2017 Guessing minutes URL: http://www.w3.org/2017/03/29-forms-minutes.html People with action items: steven[End of scribe.perl diagnostic output]