<trackbot> Date: 08 October 2014

<nvdbleek> will be running late a couple minutes

Call next week

[Regrets from me]

Steven: I'm away next week.

Erik: Let's skip next week then

Alain: Agree. I can't come either

DST

Advance warning.

Ends North America Sunday, November 2, 2014

Ends Europe Sunday October 26, 2014

Steven: The last week of October the call will be a different time for Europeans, but that is also the week of TPAC, so we may not even have a call.

TPAC meeting

http://www.w3.org/2014/11/TPAC/

Meeting 31 Oct, modulo Erik's Halloween march

Erik: That should be alright.

Review of Section 1-3

http://lists.w3.org/Archives/Public/public-forms/2014Sep/0016

Steven: I made all the changes agreed on.

Erik: I still have to make the changes for points 4 & 5 from my email.

Steven: You can do a diff to see the changes so far.

Review of Section 4-6

https://www.w3.org/MarkUp/Forms/wiki/XForms_2.0#Expressions_in_XForms

Steven: The functions are defined in Topic: Review of Section 4-6

https://www.w3.org/MarkUp/Forms/wiki/XForms_2.0#Expressions_in_XForms

Follow the second link from there

Steven: And Alain asked: "At "6.1 Model Item Property Definitions", I am not convinced that allowing

multiple type properties is effectively a good idea. For implementors, this

means multiple types per node. For authors, it might be better to explicitly

define extra types. Anyone with a use case??

"

Steven: So the question is, does having multiple types make sense? We forbid multiple calculates, sop we could also forbid multiple types.
... Erik, do you have experience, or use cases?

Erik: I don't see the purpose.
... I don't know what you would do with a conflict either

Steven: So Alain is proposing we disallow multiple types on a bind.

Erik: Types can be assigned dynamically via a bind, so we have to decide what happens when you discover it.
... stopping processing is not a good solution
... we could make this a binding error

Steven: What if the same node got bound to the same type twice, would that also be an error?

Erik: And what if two types are compatible? Such as a string and a restricted string?
... or xs:decimal and xf:decimal
... 1) we could strictly disallow, 2) allow equality or 3) allow compatibility.

Steven: Or you could require it to be validated against both types.
... one could be an int <100, and the other an int > 10

Erik: Not sure if there is much utility to allow more than one type.

Steven: And I hear Alain saying that it is hard to implement.

Erik: we could disallow it, and dispatch an error event, or we could just accept one or the other.

Steven: The current spec says it has to conform to all types

Erik: I don't know even what Orbeon does.

Steven: Let's decide this at a future call.

<scribe> ACTION: Erik to send mail listing options for the bind to multiple types problem. [recorded in http://www.w3.org/2014/10/08-forms-minutes.html#action01]

<trackbot> Created ACTION-1979 - Send mail listing options for the bind to multiple types problem. [on Erik Bruchez - due 2014-10-15].

Steven: https://www.w3.org/MarkUp/Forms/wiki/XForms_2.0#Typed_Values
... "When the expression language is XPath 2.0 or newer, or XQuery 1.0 or newer,"
... where does that come from?

Erik: This would be a hypothetical case (XQuery)

Steven: "if the expression is any other type of expression, the error is considered a dynamic error"

Erik: Make it an expression error event

Steven: "Editorial note: xs:untypedAtomic"

Erik: Remove the note

Steven: In the examples such as https://www.w3.org/MarkUp/Forms/wiki/XForms_2.0#The_type_Property
... they are too complicated, you immediately get confronted with schemas, I propose some simpler examples first.
... https://www.w3.org/MarkUp/Forms/wiki/XForms_2.0#The_required_Property
... "An XForms Processor must provide an indication that a form control is required, "
... that is a MUST I suppose

<nvdbleek> RFC 2119 to be correct

Steven: "whose expression is evaluated, converted to a string with the string conversion function as described in the expression module, and stored as the value content of the calculated data node."

https://www.w3.org/MarkUp/Forms/wiki/XForms_2.0#The_calculate_Property

Steven: is that "converted to a string with the string conversion function" necessary?

Erik: Yes.

https://www.w3.org/MarkUp/Forms/wiki/XForms_2.0#The_constraint_Property

Steven: "Note:

Specifying minimum and maximum occurrences for nodes in the instance data can be achieved by using the count() function within a constraint property."

Steven: What are we trying to say here?
... Better as an example?

Erik: Or remove entirely, preferable I think

<nvdbleek> no

Steven: Anyone else got comments on 4, 5, 6?
... Next call, section 7, quite a big section, in 2 week's time.

AOB?

[None]

[ADJOURN]

This is scribe.perl Revision: 1.138  of Date: 2013-04-25 13:59:11  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/AL/Al/
Succeeded: s/Erik,/Erik:/
Succeeded: s/ANd/And/
No ScribeNick specified.  Guessing ScribeNick: Steven_
Inferring Scribes: Steven_
Default Present: Steven_, pfennell, ebruchez, [IPcaller], nvdbleek
Present: Steven_ pfennell ebruchez [IPcaller] nvdbleek Alain
Agenda: http://lists.w3.org/Archives/Public/public-forms/2014Oct/0007
Found Date: 08 Oct 2014
Guessing minutes URL: http://www.w3.org/2014/10/08-forms-minutes.html
People with action items: erik