WebCGM WG Teleconference -- 23 Aug 2006

<trackbot> Date: 23 August 2006

<Chris> Meeting: WebCGM WG f2f, Cologne

<lofton> Scribe: Dave

<lofton> Scribenick: dc

<lofton> Scribes chosen for Session 1-6: Dave, Don, Benoit, Thierry, Chris, Stuart

Welcome, Logistics, Week's events, Review agenda

Review Goals

<Chris> The goals sound good to me

<Chris> LC= Last Call. CR = Candidate Recommendation

finalize LC processing

schedule CR spec production

finalize agreement on CR exit criteria

look at implementations and test suites

plan & agree CR/PR schedules

(start) prepare CR transition materials

Webnesday afternoon we will have a telecon with the inplementers not present

Last Call processing

<Chris> I reminded WAIPF about an outstanding review

<Chris> http://lists.w3.org/Archives/Public/public-webcgm-wg/2006Aug/0048.html

<thierry> Disposition of comments for WebCGM 2.0 Last Call

<thierry> http://www.w3.org/2006/07/03/WebCGM2-LastCallResponses.html

All last call comments received to date have been resolved

May still get comments from WAIPF

but as we move to CR processing we have a choice of how we deal with them

two loose ends from internal discussions

1. namespace defintion

2. Benoit's rewording of highlight description

namespace URI comment url

http://lists.w3.org/Archives/Public/public-webcgm-wg/2006Aug/0038.html

System identfiercurrently in WebCGM 2.0

http://docs.oasis-open.org/webcgm/v2.0/webcgm20.dtd

We seem to be ok to keep the URI the same

We seem to be ok to keep the Public and System identifier the same

The dereference document should reflect the doublicity of the specification with both W3C and OASIS

<scribe> ACTION: Chris to communicate with directory with namespace uri and identifiers [recorded in http://www.w3.org/2006/08/23-webcgm-minutes.html#action01]

<trackbot> Created ACTION-8 - Communicate with Director with namespace uri and identifiers [on Chris Lilley - due 2006-08-30].

<lofton> current NS URI: http://www.cgmopen.org/schema/webcgm/

Resolution: adopt WebCGM namespace for all versions of

Namespace URI discussion complete

Highlight method clarification original post

http://lists.w3.org/Archives/Public/public-webcgm-wg/2006Aug/0001.html

Benoit's proposal

http://lists.w3.org/Archives/Public/public-webcgm-wg/2006Aug/0044.html

BB - focus on object behavior to resolve this question

BB - if you highlight an object, does the highlight propagate to children objects?

WebCGM 2.0 reference:

http://www.w3.org/TR/2006/WD-webcgm20-20060623/WebCGM20-IC.html#webcgm_3_1_2_4

<Chris> void highlight(in WebCGMNodeList nodes,

<Chris> in boolean state);

<Chris> void highlight(in WebCGMNodeList nodes)

<Chris> void clearHilight()

<Chris> replace the first one (from the spec) with the two methods. one to add hilighting and one to clear it all

<Chris> perhaps change the method name to addHilight

<Chris> This makes the DOMmatch the object behaviors

Currently the DOM doesn't match the description of the object behaviors

Agreed that when an object is highlighted, all of it's children highlight

Resolution: Acept Chris' proposal

<scribe> ACTION: Editor to revise the document and the IDL [recorded in http://www.w3.org/2006/08/23-webcgm-minutes.html#action02]

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

<scribe> ACTION: Lofton to revise the document and the IDL [recorded in http://www.w3.org/2006/08/23-webcgm-minutes.html#action03]

<trackbot> Created ACTION-9 - Revise the document and the IDL [on Lofton Henderson - due 2006-08-30].

<scribe> ACTION: Stuart to update the highlight test in the TS to reflect this decision [recorded in http://www.w3.org/2006/08/23-webcgm-minutes.html#action04]

<trackbot> Created ACTION-10 - Update the highlight test in the TS to reflect this decision [on Stuart Galt - due 2006-08-30].

<lofton> (....morning break....)

highlight behavior discussion complete

Begin CR discusions

item left over from first previous topic

Should we process an erratum to deal with changes to URI/IRI for WebCGM 1.0?

Chris - this is probably a good idea

<Chris> http://www.w3.org/Graphics/WebCGM-19990121-errata

<Chris> sorry, thats the fe

<Chris> http://www.w3.org/Graphics/WebCGM-20011217-errata

<Chris> http://www.w3.org/TR/REC-WebCGM/

<scribe> ACTION: Lofton to produce produce the 1.0 erratum for URI/IRI clarification and circulate to Wg [recorded in http://www.w3.org/2006/08/23-webcgm-minutes.html#action05]

<trackbot> Created ACTION-11 - Produce produce the 1.0 erratum for URI/IRI clarification and circulate to WG [on Lofton Henderson - due 2006-08-30].

Timeline for CR Draft

thierry - need a minimum date for exiting CR to put into the Status of This Document (SOTD)

<tmichel> WebCGM 2.0 CR version cover page.

<tmichel> http://www.w3.org/Graphics/WebCGM/drafts/thierry-editor/overview.html

We will need to pass a resolution in the WG if we want to transition to CR

At that point we will need a conference with the director for approval to publist the CR

This conference requires a 1 week lead time

Best case - we have a Friday resolution and the document ready to go at that time

Best case conference with director would be Tuesday September 5

One month duration would be October 5

Before entering CR we would need a draft implementation report

Transition requiements for CR

http://www.w3.org/2005/08/transition?docstatus=cr-tr

*dc test

<Chris> From SVG spec:

<Chris> When two line segments meet at a sharp angle and miter joins have been specified for 'stroke-linejoin' , it is possible for the miter to extend far beyond the thickness of the line stroking the path. The 'stroke-miterlimit' imposes a limit on the ratio of the miter length to the 'stroke-width' . When the limit is exceeded, the join must be converted from a miter to a bevel.

<Chris> The limit on the ratio of the miter length to the 'stroke-width' . The value of <miterlimit> must be a number greater than or equal to 1. Any other value shall be treated as unsupported and processed as if the property had not been specified.

<Chris> The ratio of miter length (distance between the outer tip and the inner corner of the miter) to 'stroke-width' is directly related to the angle (theta) between the segments in user space by the formula:

<Chris> miterLength / stroke-width = 1 / sin ( theta / 2 )

<Chris> For example, a miter limit of 1.414 converts miters to bevels for theta less than 90 degrees, a limit of 4.0 converts them for theta less than approximately 29 degrees, and a limit of 10.0 converts them for theta less than approximately 11.5 degrees.

<Chris> http://www.w3.org/TR/SVGMobile12/painting.html#StrokeMiterlimitProperty

Initial telecon with implementors

<thierry> Pass all tests exept 6 tests:

<thierry> Interested in 2.0 test suite result, but we are now interest into 1.0 tes

<Chris> linecap 1 and 2, linejoin

<Chris> miter limit

<Chris> 2 pattern tests

<Chris> sizing of the pattern is not correct

<Chris> everyone seems to pass the values tomicrosoft gdi so they do not get the correct result

<thierry> 6 probllematic tests:

<thierry> The old 1.0 static founctionalities are the one that could raise issues. We are OK with the 2.0 tests

<thierry> Lofton: Can you send more info when you have more results ?

<Chris> SVG has some animated miterlimit tests

<Chris> animate-elem-35-t

<Chris> file://localhost/E:/cygwin/home/Chris/W3C/WWW/Graphics/SVG/Group/repository/testsuite/1.1/htmlObjectHarness/full-animate-elem-35-t.html

<thierry> Probably by tomorrow.

<thierry> Ulrich: Will send the result tomorow:

<thierry> Lofton: For the schedule we are thinking about a 30 day CR phase beginning sept 5th, ending Oct 5th

<thierry> Lofton: Any adjustement we need to do in this time frame

<thierry> Lofton we are already covered on the 2.0 tests

<thierry> Dave: we have a problem on the following funct - InterpollatedInterior

<thierry> Ulrich: We do not support this funct yet

<thierry> Dave : It does work in our 1.0 implementation.

<Chris> Larson 2.0 viewer uses OpenGL for rendering

<thierry> Don: But not as good in our 2.0 viewer

<thierry> Ulrich will send an Implementation to Chris and Lofton tomorrow

<thierry> Ulrich has now left the telecon.

CR exit criteria review

<thierry> http://lists.w3.org/Archives/Public/public-webcgm-wg/2006Jul/0084.html

<thierry> Lofton: we could change the profile, losen for 1.0 functionalities that are issues

<thierry> Chris: We can mention in the CR draft that these funct are at risk.

<thierry> Don: We are going to stuff we have resolved a long time ago

<thierry> Don: The DOM is THE objective

<thierry> Don: Each vendor want to have a good static capabilities, but vendors really want to focus on a DOM

<thierry> Chris: You can say it is required, if it is not correctly implemented

<thierry> Lofton: We are not going to remove a 1.0 implementation because we do not have two perfert implementation

<thierry> Chris: it is better to say in the Profile that it is optional than to have diffrent rendering in different viewers

<thierry> Chris: we could have also some authoring guidelines

<thierry> Benoit: Would be happy with relaxing the profile

<thierry> Don: We could implement the Minorlimit

<thierry> Don: we are 2 implementations; one for static and a new one handling Dynamic

<thierry> Don: the real interoperability should be the DOM

<thierry> Benoit: If chris says the only way is to change the profile, I am fine with that

<thierry> Chris: W3C should probably have a sort of QA phase were we could use much more tests to test implementations

<thierry> Chris: These would not be substantive changes, needing another Last Call

<thierry> Lofton: to pass the patern tests we should have the size optional ?

<thierry> Don: The gradient in the Interpolated Interior is the issue we have

<thierry> Lofton: Need to be clear on which tests we have problems with

<thierry> Lofton: Either two passes on each tests or downgrade the feature

<thierry> Lofton: the feature at risk wil be the Interpolated Interior

<thierry> Benoit: The miter limit we could have that done in our implementation.

<thierry> Benoit: If we have time we should look at the impact on the profile

<thierry> Benoit: I was thinking about miter limit, to to say so in the profile

<thierry> Benoit: there is also an internal issue raised by Ulrich (Benoit will repport about that)

<thierry> esolution: the candidate feature at risk are the pattern size aspect of deniable patterns and the performance requirements for miter limit

<thierry> Resolution: the candidate feature at risk are the pattern size aspect of deniable patterns and the performance requirements for miter limit

<thierry> Lofton: the feature Interpolated Interior is covered by the Larson 1.0 Implementation, so we are OK with this one

<thierry> Benoit: we are pretty good on Unicode

<thierry> Lofton: In the Test suite, there are 7 tests that are Chinese and Japanese, where two are IRI tests using Unicode caracters

<thierry> Dieter : Pattern size - It is cearly a bug. We will fix it.

<thierry> Dieter: miter limit: We think we pass this test.

<thierry> Chris: It depends if you are implementing a converter or a viewer

<thierry> Dieter: It is only a screen issue but it is correctly printed on a PS printer

<thierry> Dieter: actually a screen performance issue.

<thierry> Dieter is fine with the DOM change about highlight resolution discussed this morning

<thierry> Benoit: we would be inrested in a flash object method

<thierry> Lofton: that would be for 2.1 ...

<thierry> Lofton: we have preliminary features for 2.1 and 3.0

<thierry> Dieter: I would like to make sure this can be added latter on

<thierry> Lofton: At this point I would like to have the minimum changes, this is the case for the highlight methodFlashing

<thierry> Don: I would also like to wait before adding it. It may be tight to other features

<thierry> Dieter: We could replace true/flase with flash On/off in the clear hightlight method

<thierry> CHris: instead or True/false we could use a numeric value, allowing to add another value latter on, not a bolean value.

<thierry> Dieter: Yes that is a good suggestion

<thierry> Lofton: Sounds fine with me.

<thierry> Resolution: second parameter is called behavior, and the value is add

(continue) CR process discussions

<thierry> Lofton: will exit CR on 5th October

<thierry> Lofton: This should allow to more to PR by end of October

<thierry> Thierry: yes before Ac meeting in November. it would be usefull to advertize during the AC meeting wehere all AC Reps are present

<thierry> http://www.w3.org/Graphics/WebCGM/drafts/thierry-editor/overview.html

<thierry> Thierry: The Cover page is ready. It does incorporate the publication date on sept 05. Ending CR date on oct 5th

<thierry> Thierry : SOTD includes the exit criteria adopted by the WG and 3 features at risk

<thierry> features at risk are : Pattern size aspect of definable patterns, Performance requirements for Miter limit., Text on a path

<thierry> The cover page also fulfills the pubrules

<thierry> Thierry: Lofton needs to copy and paste it into the CR draft

<thierry> meeting ajourned

- DRAFT -

WebCGM WG Teleconference

23 Aug 2006

Attendees

Contents