<artb> Date: 18 April 2007

<Hixie> sure

<artb> hixie - can you join the W3C bridge?

<Hixie> which code?

<artb> 9231

<artb> yeah, whatever, zakim :)

<Hixie> what's the phone number again?

<Hixie> been so long i can't remember it anymore :-)

<artb> +1 617 761 6200

<Hixie> thanks

<marcos> scribe: Marcos

scribnick: marcos

XBL 2

http://lists.w3.org/Archives/Member/member-appformats/2007Apr/0002.html

Artb: I wanted to go through the three agenda items regarding to XBL
... I remember Jonas had made some comments about XBL.

Hixie: We havent recieved many comments yet, I would like to wait for more implementation experience and comments to come in and

then address them all at once.

Artb: Based on the comments, Will we have to go back to last call?

Hixie: we might have to go back to last call.

Artb: It sounds reasonable that we might have to go back to last call to address the comments.

Cam: it might be nice to start addressing some of the emails now.

Artb: what are your thoughts on going ahead and start responding to Jonas' comments, it might be good to keep the public list

up-to-date.

Arb: where there any other comments apart from Jonas?

hixie: there were a

few comments from Anne, etc, but few other significant ones.. Jonas seems to be the only mayor person.

<xmarcos2> Artb: I'd like to get a general idea about how we are going to do the test suite

<xmarcos2> Hixie: I have 3 suggestions. It would be very helpful if the test suite followed the guidelines simply say if you have passed or failed within one second.

<Hixie> http://hixie.ch/tests/evil/css/css21/guidelines/

<xmarcos2> CSS CSS WG's guidelines for test suites

<Hixie> http://cgi.w3.org/member-bin/process.cgi?method=url&url=http%3A%2F%2Fwww.hixie.ch%2Ftests%2Fevil%2Fcss%2Fcss21%2Fguidelines%2Fguidelines.src&output=html

<xmarcos2> Hixie: those guidelines are quite useful. Hopefully we will get a large number of tests. The best way of getting lots of test is to open them up to the public. We are going to get lots of tests, so it's going to be hard to check them all so we might assume that they are valid until they are found to be otherwise by implementers.

<xmarcos2> Artb: does anyone have any test suite experience they would like to share?

<xmarcos2> Artb: regarding getting contributions from public, we need to find out if there are any W3C related issues (copyright). However, Ian, how do you think we could make this work?

<xmarcos2> Hixie: We might need to appoint a point man. The thing is to start requesting tests. Implementer will probably contribute tests too as they need them regardless.

<xmarcos2> Artb: I guess I would like to get some commitment on the creation of test from people from the Wg

<xmarcos2> Anne: I might contribute, but not actively until there is an implementation.

<anne5> (and even then I might not)

<Hixie> yeah most people won't contribute until there are impls

<Hixie> that's why i think impls will have to come first, and provide the first tests

<artb> Marcos: Lachy and I have played around with Mikko's impl

<artb> ... it isn't complete but it has been helpful in the context of the Primer

<xmarcos2> Artb: can I get a volunteer to think about the test suite template?

<artb> ACTION: Cam and Lachy to create a test case template. [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action01]

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

<artb> ACTION: McCormack work with Lachlan to create a test case template that makes sense for XBL2 test cases. [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action02]

<trackbot> Created ACTION-81 - Work with Lachlan to create a test case template that makes sense for XBL2 test cases. [on Cameron McCormack - due 2007-04-25].

<xmarcos2> Hixie: the problem with XBL is that there are so many different ways of embedding it into different documents.

<xmarcos2> Hixie: I think the best way to do the test suite would be to target different types of documents to cover a wide range of test cases.

<xmarcos2> Artb: we could be looking at hundreds if not thousands of tests.

<xmarcos2> CM: yes, the problem being that there is no one host language for XBL.

<xmarcos2> CM: it's might be best to just start some simple test cases in HTML as that might be the biggest use case for XBL.

<xmarcos2> Artb: does anyone have anything else about the test suite?

<xmarcos2> none...

<xmarcos2> Artb: lets talk about the potential implementations. We took a sample of potential implementers three months ago so it might be good to take another sample now.

<xmarcos2> Mikko, you around?

<artb> ACTION: Barstow contact Mikko to determine HUT implemenation plan for their JS-based XBL2 implementation [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action03]

<trackbot> Created ACTION-82 - Contact Mikko to determine HUT implemenation plan for their JS-based XBL2 implementation [on Arthur Barstow - due 2007-04-25].

<xmarcos2> Artb: CM, have you got any plans to implement anything?

<heycam> I have vague plans to convert the sXBL implementation in Batik to XBL 2.0, which maybe I'll do around the end of the year. It's not high on my list of priorities though.

Artb: does nokia have any plans?

GG: not at this point.

Artb: Anne, any plans from Opera?

Anne: no comment

Artb: how about you Marc? is Microsoft interested in implementing?

MS: I really can't comment on if an when, but we are interested in it.

Marc, it's 30c today :D

Artb: Hixie, do you know if Mozilla is interested in implemented?

Hixie: they have the same position as microsoft...

<artb> ACTION: Barstow ping Apple about their XBL2 implementation plans. [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action04]

<trackbot> Created ACTION-83 - Ping Apple about their XBL2 implementation plans. [on Arthur Barstow - due 2007-04-25].

Artb: Apple joined the WG recently, but we are unsure if they are interested in XBL
... have you chatted with David Hyatt at all about XBL and gotten any comments?

Hixie: yes, Apple said it needed further reviewing

Artb: When is the candidate expiration period? Somewhere in September I think. However any chance of getting out of candidate by then is probably 0. Do you know what the effect of that might be in relation to the w3c?

XBL2 Primer

<anne5> scribenick: anne5

MC: So...
... ... I contacted LH
... See http://lists.w3.org/Archives/Public/public-appformats/2007Apr/0009
... Need to look at use cases on the web.
... Content reordering will likely be common
... Start the primer with that... demonstrates bindings
... should be simple enough and useful

GG: XBL might have use with Widgets. Making Ajax frameworks easier to use

MC: you have web development and there's web applications stuff like the Web 2.0 movement, libraries, blah
... start simple, move to app level
... [elaborates on the above e-mail]

AB: Why is reordering important?

MC: You might want to reorder content for mobile devices. etc.

GG: Does the author need to be involved?

AvK: yes

GG: CSS might be feasible as well

AvK: CSS doesn't allow reordering of actual elements though, like XBL

GG: Ah right, changing the tree-shape

[Some talk about how to use SVG to style elements through XBL, color pickers, sliders, progress bars, etc.]

<heycam> ACTION: marcos to investigate a colour picker widget example (e.g. binding <input type=colorpicker>) for the xbl primer [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action05]

<trackbot> Created ACTION-84 - Investigate a colour picker widget example (e.g. binding <input type=colorpicker>) for the xbl primer [on Marcos Caceres - due 2007-04-25].

<Lachy> could create an XBL reference like these http://www.ilovejackdaniels.com/cheat-sheets/

[Some discussion on XBL2 and HTML5.]

[And Web Forms 2]

<artb> http://www.w3.org/Member/Mail/AuditForm

<artb> ACTION: Marcos and Lachlan to have XBL2 Primer FPWD ready by July f2f meeting. [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action06]

<trackbot> Created ACTION-85 - And Lachlan to have XBL2 Primer FPWD ready by July f2f meeting. [on Marcos Caceres - due 2007-04-25].

<jcantera> yes

<jcantera> are you mutted¿

<jcantera> ok

<scribe> scribenick: anne

<anne5> scribenick: anne5

AB: [explains history of web app delivery]
... charter expects a XUL-like language deliverable
... pushed by Nexaweb + Telefonica
... rest of the WG felt like the work should be about improving existing work as opposed to doing something new
... progress has been slow
... HTML, XHTML2 or DIAL might be better to address this stuff

JC: 1. we recognize that this work is only driven by two companies in the WAF WG
... has been very difficult
... two companies is not enough
... 2. more companies should be involved

AB: do you have data that suggests that other members will actively participate in this work?

JC: not at the moment

charter update

AB: we need a charter update to more accurately reflect what we're doing
... Since only 2 of the 400 members are interested in the DFAUI deliverable we should really think about it

JC: I agree that an XG might be a solution
... for the DFAUI work
... I don't think it's an issue that only a couple of companies are involved.

<scribe> scribenick: marcos

AB: we want a charter that accurately reflects what we are doing. I think we should use the rechartering process to work out how to move the DFAUI forward. The good thing about doing this work in an incubator is that it provides a great deal of flexibility (ie, select your own chair, set your own timeframe).
... I guess it would be good for you guys to continue to look how specs at the w3c meet the use case and requirements of DFAUI to identify gaps.

JC: I don't see as realistic progressing the work in relation to HTML5 as HTML might be too basic for what we need.

MC: I don't see any use cases that can't be addressed by HTML5. I would like to see concrete evidence that HTML 5 is not already a DFAUI.

Artb: I'm suggesting that the DFAUI work move to an incubator group.

JC: I agree that that could be a solution we could explore.
... art explains what is the process of an incubator group as he understands it...

Artb: the work under the incubator group could then delegate how the actual DFAUI could be progressed by exploring further what other working groups are doing and which groups could potentially meet the requirements of the DFAUI
... art continues to explain what an Incubator group...

<scribe> ACTION: Arthur Barstrow to talk to Chris Lilly about how we go through rechartering [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action07]

<trackbot> Created ACTION-86 - Barstrow to talk to Chris Lilly about how we go through rechartering [on Arthur Barstow - due 2007-04-25].

Artb: I've had to report to W3C management continuously that there has been no progress on the DFAUI and this may soon become a concern for the w3c.

JC: we should not make any decisions without Coach here.

AB: agreed
... I need to speak to Chris Lilly about the rechartering process and set up a teleconf so we can discuss how to best more this work forward
... without coach, my recommendation is that we don't review the DFAUI use case and reqs document

JC: agreed
... What do you need to start an incubator group?

<artb> Incubator Group Activity: http://www.w3.org/2005/Incubator/

AB: does anyone else have anything else regarding this topic?

JC: I am a bit concerned about the timing.

AB: I will speak to Chris about the timing and get back to you.

<artb> Topic XBL2 Primer

<artb> JC: what is the status of the Primer?

<artb> MC: Lachlan and I have some work to do.

<artb> ... We do not expect a FPWD until July or so

<Coach> anyone on?

Coach?

<anne5> DFAUI discussion just sort of ended

<Coach> yes.

<Coach> ok.

<Coach> now i can go back to sleep now.

nn

:)

<artb> scribe: Lachy

GG: we've talked about the need for digitally signing widgets
... proposal is based on XML Sig
... Solution is capable of signing parts of a widget
... May or may not be needed
... [shows examples]
... more than one signature possible. e.g. author and distributor could each sign it separately
... is that the way forward? If not, we may need another solution

CM: Has it been compared with other signing techniques? e.g. Java JAR files

MC: [describes Yahoo's technique]
... This is just one option, there are other possibilities
... currently having problems with signature authorities

<Coach> isn't there going to be a discussion around DFAUI now (from 2am ET)?

<anne5> It ended pretty much the first time you joined.

<anne5> Jose was online back then as well and it was discussed with him.

<anne5> We're now discussing signatures for Widgets.

<Coach> oh...i thought it should start at 2am. and i was you were joking.

<Coach> i thought you were joking.

<anne5> I was not.

[discussion of certificate authorities and public keys, and how they would work with widgets]

Coach: it was supposed to start at 14:00 local time here in Brisbane, though it started at about 14:45

<anne5> And it's now 16:47 local time fwiw.

MC: We should probably recommend support for X509

AvK: xml:id broke XML Canonicalisation
... XML Canoicalisation is being fixed

MC: Use case for sig is that users need to be able to verify the distributor

CM: Sigs have benefit of integrity checking

MC: This may not be as difficult as Yahoo!'s solution
... Possibility that some things may not be signed

GG: Should probably require everything to be signed

AB: Any objections to this proposal?

AvK: Integrating this and getting it to work may be difficult

<scribe> ACTION: Complete the proposal for widget signing [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action08]

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

<scribe> ACTION: Guido Complete the proposal for widget signing [recorded in http://www.w3.org/2007/04/18-waf-minutes.html#action09]

<trackbot> Created ACTION-87 - Complete the proposal for widget signing [on Guido Grassel - due 2007-04-25].

AB: Let's diiscuss the Widgets spec and pick up where we left off

AvK: The way the security model works has implications on how other things work

<anne5> http://dev.w3.org/cvsweb/~checkout~/2006/waf/widgets/Overview.html?content-type=text/html;%20charset=utf-8

<marcos> http://dev.w3.org/cvsweb/2006/waf/widgets/

<marcos> http://dev.w3.org/cvsweb/~checkout~/2006/waf/widgets/Overview.src.html

AvK: Probably want to be able to specify access rights and restrictions

GG: Widgets can't have access to other widgets or the browser
... Should be able to white/black listing
... of domains

AvK: a widget is just a special type of application, why should it have restrictions beyond those of other apps?

GG: Widgets may request permission from users

AvK: Users will just say ok without understanding the question
... Would be ok to be able to turn on/off network and/or file access
... [asks about Firefox Extensions security model]

MC: White lists may just list malicious sites

LH: author specified white lists can't work securely

MC: UIs for requesting access from the user are confusing

[current discussion being recorded in the issue list in the draft]

[Discussing Widget Geometry]

AvK: The WebAPI WG is specifying the Window object

AB: Can we move the widget geometry to another spec?
... like the Window API spec

AvK: Could move it to HTML5

[Discussion about whether or not we should require widgets to use HTML]

AvK: We have to define how HTML works in a widget
... HTML and SVG will most likely be the dominant formats

MC: Yahoo doesn't use HTML for their widgets

AvK: We should definately use rel=widget
... The type attribute shouldn't be required

MC: Why reference HTML5 instead of HTML4?

LH: We could get away with referencing HTML4 for the link element

<artb> Anne's input: http://lists.w3.org/Archives/Member/member-appformats/2007Apr/0010.html

http://lists.w3.org/Archives/Member/member-appformats/2007Apr/0010.html

http://lists.w3.org/Archives/Member/member-appformats/2007Apr/0011.html

[Anne is describing the proposal from Lachy and himself]

<heycam> CM: rename description to desc, get its default value from the desc element if it's an svg main file

[some dispute about the proposed <index> element name]

MC and AB like the idea of <main>

AvK: <start>?

LH: <default>?

<anne5> I've proposed <init>, <start> and <index>

AB: doesn't like <index>

LH doesn't like naming debates

<anne5> I don't like <default>

AvK: I can't spell main

<anne5> main is non-obvious and doesn't make much sense here

<artb> RESOLUTION: use <start> rather than <index>

This is scribe.perl Revision: 1.128  of Date: 2007/02/23 21:38:13  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/guidelines/CSS WG's guidelines/
Succeeded: s/MC:/CM:/
Succeeded: s/GR:/GG:/
Succeeded: s/Mark?/Marc?/
Succeeded: s/might better/might be better to/
Succeeded: s/about i/about it/
Succeeded: s/fil/file/
Found Scribe: Marcos
Inferring ScribeNick: marcos
Found ScribeNick: anne5
Found ScribeNick: anne
WARNING: No scribe lines found matching ScribeNick pattern: <anne> ...
Found ScribeNick: anne5
Found ScribeNick: marcos
Found Scribe: Lachy
Inferring ScribeNick: Lachy
Scribes: Marcos, Lachy
ScribeNicks: marcos, anne5, anne, Lachy
Default Present: Ian_Hickson

WARNING: Replacing previous Present list. (Old list: Art, Cam, Guido, Anne, Marcos, Lachlan)
Use 'Present+ ... ' if you meant to add people without replacing the list,
such as: <dbooth> Present+ Art, Anne, Guido, Cameron, Lachlan, Marcos, Marc_(XBL2), Hixie_(XBL2)

Present: Art Anne Guido Cameron Lachlan Marcos Marc_(XBL2) Hixie_(XBL2) Jose_(DFAUI)
Agenda: http://lists.w3.org/Archives/Member/member-appformats/2007Apr/0002.html
Found Date: 18 Apr 2007
Guessing minutes URL: http://www.w3.org/2007/04/18-waf-minutes.html
People with action items: arthur barstow barstrow cam complete contact guido lachlan lachy marcos mccormack mikko with work