See also: IRC log
[USA: 12 March; Europe: 26 March; so for two weeks call is 1 hour earlier
in Europe]
Steven: CFP, deadline is the end of this month
https://lists.w3.org/Archives/Public/public-xformsusers/2017Feb/0023.html
Steven: Any feedback, or are you happy?
Philip: Still to do.
Steven: It's quite short, by merging sections it has compressed quite a lot.
https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0000.html
Steven: Isn't this already covered?
Erik: I think we say something
about the use of these cross-MIP functions
... let me see if I can find it.
<ebruchez> https://www.w3.org/community/xformsusers/wiki/XPath_Expressions_Module#The_valid.28.29_Function
Erik: I think the second example
in my mail wouldn't work at present
... but it would if recalculate and validate were combined
Steven: I see.
... To what extent is the way we specify the processing in
these distinct steps essential?
... As I mentioned last week, there are ways of combining
rebuild and recalculate
... and you are suggesting ways to combine recalculate and
revalidate
Erik: We are a bit too
prescriptive in the specification
... though the cross-MIP functions make it harder
Steven: I moved a line about being able to optimise as long as the result is the same (a bit redundant really) from refresh to the top of the model processing:
"Although some processes in this specification are described as if all expressions in a model or the user interface are re-evaluated every time there is a change, implementations are of course at liberty to optimize this, as long as the result is the same."
https://www.w3.org/community/xformsusers/wiki/XForms_2.0#Dependencies
Steven: So my question is, to what extent is it essential that we are prescriptive about the RRRR steps.
Erik: Is there anything that
prevents us from doing it in a different way?
... the functions could create a case where you get a different
result depending on whether you do them as four steps rather
than less.
... I can see why it was written how it was.
Steven: What I want to avoid is
that we prevent an implementation doing the best thing, by
using wording in the spec that requires them to do something
else.
... there's no reason why a really good dependency algorithm
couldn't do RRRR in one go.
Erik: Refresh may need to be separate.
Steven: So what do we do with the RRR actions? What do we specify that they do in the case an implementation with a monolithic dependency algorithm?
Erik: We say that the three are the same.
Steven: So we say that <revalidate/> does *at least* a revalidation? That sounds good.
https://lists.w3.org/Archives/Public/public-xformsusers/2017Feb/0022.html
Steven: I haven't worked on this
this week.
... so let's keep it on the agenda.
https://lists.w3.org/Archives/Public/public-xformsusers/2017Feb/0026.html
Steven: I'm not objecting to <dialog/> per se, just that it isn't very well integrated into the whole
<ebruchez> https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0001.html
Steven: We add a lot of functionality in dialog that is special just for that element, and I would prefer to see a more integrated whole.
Erik: Let's think a little bit more about it.
Steven: Very good.
[None]
[Adjourn]
<scribe> ACTION: Steven to change definition of RRR actions to say that they do *at least* R [recorded in http://www.w3.org/2017/03/01-forms-minutes.html#action01]
<trackbot> Created ACTION-2112 - Change definition of rrr actions to say that they do *at least* r [on Steven Pemberton - due 2017-03-08].
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/ee// Succeeded: s/pre/per/ Succeeded: s/a buit/a bit/ Succeeded: s/comperssed/compressed/ Succeeded: s/Steeven: I s/Steven: I see/ Present: Alain Erik Philip Steven No ScribeNick specified. Guessing ScribeNick: Steven Inferring Scribes: Steven Agenda: https://lists.w3.org/Archives/Public/public-xformsusers/2017Mar/0002 Found Date: 01 Mar 2017 Guessing minutes URL: http://www.w3.org/2017/03/01-forms-minutes.html People with action items: steven[End of scribe.perl diagnostic output]