Forms Working Group Teleconference

23 Jun 2010

See also: IRC log


Leigh_Klotz, wiecha, Nick_van_den_Bleeken, Philip Fennell, Erik Bruchez
Leigh Klotz


<trackbot> Date: 23 June 2010

<pfennell> And the conference call code is?

<pfennell> Thank you :-)


<scribe> Scribe: Wiecha

Vacation schedule

Leigh: pls be sure to fill out
... regrets for next three weeks

next week 2 out on vacation

Uli and Leigh

<pfennell> Sorry, I'm still having trouble dialing in, I'll be with you in a minute, or two...

<ebruchez> all, I will join in a few minutes

<pfennell> It's all going wrong for me today, either I get disconnected or it won't accept the passcode : 93676

yes, that's the code

<pfennell> I had this problem last week but it let me in eventually. No joy yet today.



converts to XML format

can pipe through XSLT to spec XML

need conventions to use in wiki markup -- references etc

<scribe> ACTION: Charlie Wiecha to propose subset of spec XML to use in the wiki [recorded in http://www.w3.org/2010/06/23-forms-minutes.html#action01]

<trackbot> Sorry, couldn't find user - Charlie

Charlie: think we want to be fairly limited in the profile of spec xml to cover

Leigh: right, just start by defining the scope
... could use XML directly in the wiki

Charlie: probably wouldn't be nice w/o validation



Leigh: reporting on status from our own use for control definitions

we're not using the parameter enhancements but taking XBL as template then generating a transcoded XBL to include parameters directly

have asked to look at Orbeon extensions

in any event our work is not specific to XForms

Charlie: so is XBL a design-time thing?

Leigh: yes, as a macro language

Erik: XBL has template language, more limited than XSLT, but is live

so if the DOM mutates then changes are applied incrementally

also provides visibility control - shadow DOMs and mapping between outside and shadow tree

communicate through events

so not just a matter of expanding the templates

we have tried to use only XSLT for reusable components

this template-only approach has limitations, e.g. author of component should be able to use IDs which are not visible

some aspects of the component should be private, encapsulated

Erik: we've had to go further than the XBL2 spec to support private visibility and another level not in the spec which allows control to surface some markup into the outer scope

Charlie: I think there might be something in the spec about content merging
... might schedule some time at the F2F to talk about Forms WG relationship to XBL2

Leigh: would be good to have a profile of XBL2 we can use

need something simpler and understandable

Erik: not sure XBL2 is going anywhere...better than XBL1 but has quirks, poor template language

doesn't help us with features more proper to XForms

e.g. create a component like xf:input but has these new features

need to reproduce aspects of the original control

need for lots of coding in the control, variables, possibly local instance

lots of boilerplate code that's required...we may have some patterns from a forms perspective that we could build on commonly

Leigh: through standard libraries?

Erik: maybe using XBL inheritance

could provide standard base XForms binding or series of them which define our behavior

then extend those for the various controls and other elements

Charlie: I'm doing something similar in Dojo using inheritance and mixins to existing Dojo form widgets

Leigh: I think it's promising approach for components
... we should publish at least a Note or better a Rec-track document on this

at least from a templating approach...i.e. not requiring XBL support on the client

Charlie: not sure it's enough to do this w/o runtime support on the client

Leigh: think there's value short of full XSLT but requires minimum runtime support

Mozilla css matching rules is how XForms loads

Erik reports that this runtime matching is not required

Erik: the fact we're not currently doing dynamic expansion doesn't mean we're not interested

important to be able to change control appearances after the page is loaded, but our first requirement is for reusable components

so reusability was first requirement and then second use runtime type or @appearance to pick proper binding

but just having a component model for static templating is useful by itself

server vs. client is an implementation detail...could do it on the client

Leigh: think there are two levels here

would like to get our experience to date written up

runtime behavior fits well with XForms behavior but is separable issue

e.g. two levels of compliance or spec

<klotz> two ways of doing output bound to date:

<klotz> 1. xforms: http://en.wikibooks.org/wiki/XForms/Formatting_a_date

<klotz> 2. xforms+xbl1: http://en.wikibooks.org/wiki/Talk:XForms/Formatting_a_date

Erik: dynamic mapping could also be a function of CSS classes
... attribute setting also propagates to attributes and content inside the shadow tree

<scribe> ACTION: Charlie Wiecha to look at JS implementations of XBL2 [recorded in http://www.w3.org/2010/06/23-forms-minutes.html#action02]

<trackbot> Sorry, couldn't find user - Charlie

<nick> http://code.google.com/p/xbl2/

Leigh: next steps: document experience from Erik, Charlie to fill in requirements from Dojo widget component model

Charlie: should tee this up for F2F discussion
... or sooner :-)

<nick> we could have a longer call on XBL2

<klotz> +1


Summary of Action Items

[NEW] ACTION: Charlie Wiecha to look at JS implementations of XBL2 [recorded in http://www.w3.org/2010/06/23-forms-minutes.html#action02]
[NEW] ACTION: Charlie Wiecha to propose subset of spec XML to use in the wiki [recorded in http://www.w3.org/2010/06/23-forms-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/06/23 16:05:19 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/XBLw/XBL2/
Found Scribe: Wiecha
Inferring ScribeNick: wiecha
Default Present: Leigh_Klotz, [IBM], wiecha, Nick_van_den_Bleeken, +44.782.483.aaaa, +1.650.515.aabb
Present: Leigh_Klotz [IBM] wiecha Nick_van_den_Bleeken +44.782.483.aaaa +1.650.515.aabb

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 23 Jun 2010
Guessing minutes URL: http://www.w3.org/2010/06/23-forms-minutes.html
People with action items: charlie wiecha

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

[End of scribe.perl diagnostic output]