SVG Working Group Teleconference -- 05 May 2011

<trackbot> Date: 05 May 2011

<scribe> Scribe: vhardy

<shepazu> Scribenick: vhardy

heycam: First topic: 1.1 publication.

shepazu: the most important thing is that we resolved to proceed with the publication.

heycam: we should make that resolution now.

shepazu: I have not seen a formal resolution to proceed. We should take one now.

heycam: do we have a resolution to transition the SVG Full 1.1 2nd Edition to proposed recommendation?

no objections.

RESOLUTION: The WG will transition SVG Full 1.1 2nd Edition to proposed recommendation

heycam: in terms about last steps we need to take...
... spec. editing is all done, except for the issue about whether or not filters should apply to mask. Erik wanted to get that in. All the previous outstanding have been addressed.
... the PR template has been applied to the spec.

<heycam> http://dev.w3.org/SVG/profiles/1.1F2/publish/

heycam: the spec. is fine. The implementation matrix is where we have some work left.

<heycam> http://dev.w3.org/SVG/profiles/1.1F2/test/status/implementation_matrix.html

heycam: I ran it through FF and WebKit yesterday. We had discussed last week about the remaining tests that do not have 2 passes and for which we decided to not use ASV. We had a discussion about unapproving these tests.
... I had a conversation with shepazu about unapproving the tests that do not have two passes. Last week, we did not specifically decide. I had an action to discuss it with shepazu to decide if it was the best way forward.

shepazu: I personally believe that it is better to take these tests into SVG 2 and not include them into SVG 1.1.

heycam: given that some aspects of the spec. are not covered to that extent, we are not weakening the test suite by unapproving these tests.

general agreement about unapproving the tests as summarized by Erik.

<heycam> http://www.w3.org/mid/op.vuyp2zyvgeuyw5@localhost.localdomain

heycam: we are down to 4 tests that are going to be unapproved and 1 that only has one green slots and references ASF for the second impl.
... if there is no objection, I'll make the changes to the impl. report after the call.

(no objection).

heycam: shepazu, should we publish the test suite differently?

shepazu: there is no formal procedures around test suite. There are best practices, but not officiel process to follow to proceed. We need to demonstrate that there are interoperable implementations, and we use a test suite for that demonstration.

heycam: we should publish it.

vhardy: I think it would be good to publish the test suite at the same time we go to PR.

heycam/shepazu: agreed.

<scribe> ACTION: heycam to publish the test suite and implementation report at the same time SVG 1.1 2nd edition goes to PR. [recorded in http://www.w3.org/2011/05/05-svg-minutes.html#action01]

<trackbot> Created ACTION-3038 - Publish the test suite and implementation report at the same time SVG 1.1 2nd edition goes to PR. [on Cameron McCormack - due 2011-05-12].

heycam: Tbah, there is a test shapes-rect-03.svg: can you test Inkscape so that I can update the impl report?

tbah: yes, I can do that.

<scribe> ACTION: tbah to verify shapes-rect-03, filters-image-04 and filters-image-05 against Inkscape. [recorded in http://www.w3.org/2011/05/05-svg-minutes.html#action02]

<trackbot> Created ACTION-3039 - Verify shapes-rect-03, filters-image-04 and filters-image-05 against Inkscape. [on Tavmjong Bah - due 2011-05-12].

heycam: that is all about the impl. report.
... the other part is the disposition of comments.

<heycam> http://www.w3.org/2010/09/SVG1.1SE-LastCall/

heycam: there were only a couple cells to fill in.
... there are only 4 comments for which there was no response to our response.

vhardy: have we sent repeated requests?

heycam: yes, and still did not get answers.
... In any way, these comments have been accepted and the changes made.
... admin-wise, in terms of spec., that is all.
... there are two things on the agenda: animate-elem-92, the other is filters applying to masks.
... the animation test is for the recent change we made to discrete 'to' animations. It would be good to include it. It passes two implementations.
... we should add it to the impl. report.

vhardy: sounds good.

(no objection).

heycam: ok, I will add animate-elem-92 to the test suite and impl. report.

<scribe> ACTION: heycam to add animate-elem-92 to the test suite and impl. report for SVG 1.1 Full 2nd edition. [recorded in http://www.w3.org/2011/05/05-svg-minutes.html#action03]

<trackbot> Created ACTION-3040 - Add animate-elem-92 to the test suite and impl. report for SVG 1.1 Full 2nd edition. [on Cameron McCormack - due 2011-05-12].

heycam: about the filter issue (applying to masks). We have not discussed this on a call. I did respond to Erik.

<heycam> http://www.w3.org/mid/op.vttgfpltgeuyw5@localhost

heycam: it seems that the spec. says filters should apply to the mask element, but no implementation does it.
... I would propose that filter does not apply to mask.
... like the recent clarification that opacity does not apply to the mask element.

(no objection)

heycam: it is a small change in the spec., like removing one word or so.

RESOLUTION: The group agrees that the filter property does not apply to the mask element.

vhardy: just to clarify, this is just on the <mask> element, you could still filter the children of a <mask>, right?

heycam: yes, that's right.
... I wrote a test for this today.

<heycam> http://dev.w3.org/SVG/profiles/1.1F2/test/svg/masking-filter-01-f.svg

heycam: I think we should add this to the test suite since all implementations should pass it. The test requires review and approval.
... can someone review it during the call? It is a simple test.
... also, let's discuss it at the end of the call.

vhardy: I agree we should add the test to the test suite.

heycam: ok, that is all the discussions we needed for 1.1. shepazu has put in motion the request for the transition.
... there is a minimum of 6 weeks before PR and recommendation?

shepazu: yes, I believe this is the case.

vhardy: what is the date of the transition call?

shepazu: the publication will probably happen on the 12th.
... the transition call could happen Friday May 6th or Monday 9th or Tuesday 10th.

heycam: the deadline we had was that we wanted the publication request to be in before Friday May 6th before the publishing moratorium.
... now that we have a resolution, we are all set up for the transition request.

F2F venue

heycam: last time we discussed this, we were wondering how to maximize the FX task force participation. Shepazu was going to talk to Patrick and Dean.

shepazu: I sent emails. Dean said unlikely. Patrick said no, but Sylvain could join instead. Sylvain said he could join for the FX day, but not the SVG meeting.
... it seems we would not have Apple or Microsoft at the meeting.

heycam: cabanier and vhardy are interested in joining.

vhardy/cabanier: we are flexible, we could attend at a different time/location.

shepazu: for that date, it is unlikely to get Microsoft, and impossible to get Apple.

vhardy: do we know of a time/location that would work for them?

shepazu: Redmond is good for Patrick.
... Seatle should work to.
... I think Microsoft might be willing to host.

cabanier: Adobe has an office too.
... I can look into hosting as well.

shepazu: it would be good to meet with the CSS working group in Japan. But it is unlikely to get Patrick, Chris and Dean.

heycam: Tab will be in Japan.
... and he participates in FX discussions.

anthony: I will not be in Japan.

shepazu: who would be in the SVG WG meeting?

Jun, Rik, Vincent (2 days), Doug, Cameron, Tab (?)

shepazu: we would not have critical mass really.
... and missing critical participants.

(Dean, Patrick, Chris, Anthony).

shepazu: I am not sure it is worth it.

vhardy: could we have a poll to decide on the next F2F.

<scribe> ACTION: shepazu to set up a poll to have two options: set date for the Japan meeting (right after CSS WG F2F) and a meeting in Redmond/Seatle with a few dates. [recorded in http://www.w3.org/2011/05/05-svg-minutes.html#action04]

<trackbot> Created ACTION-3041 - Set up a poll to have two options: set date for the Japan meeting (right after CSS WG F2F) and a meeting in Redmond/Seatle with a few dates. [on Doug Schepers - due 2011-05-12].

heycam: We could also have F2F co-located with the TPAC (November, Boston). SVG Open is in Boston late September).

shepazu: I think TPAC is in California, not Boston.

<heycam> SVG Open is October 17 to 20, 2011

<shepazu> TPAC: Santa Clara Marriott, Santa Clara, California, (Silicon Valley) USA

<shepazu> 31 October to 4 November 2011

heycam: July is still a reasonable date for the group.

shepazu: late July would work for me.

vhardy: for me too.

heycam: so the options are a) Japan early June and b) Seatle late July.

shepazu: yes.

heycam: I hope we can get the poll sorted out quickly. Jun is waiting to hear from us.

vhardy: suggest that we could beef up our test suite.

shepazu: we could also start out on SVG 2. But I do not know what Cameron's travel restrictions are.

(discussion on whether or not would be too much travel).

Discussion on SVG 2. Based on J Watt's email.

anthony: The test masking-filter-01.svg looks good except for the test description which needs a pass criteria.
... I would also slightly modify the test to move the green rectangle outside of the group so that is it always guaranteed to show up on top of the red.

(discussion about the test)

heycam: I'll update the test and add a sub-test.

anthony: having a mask on the group applies to the whole group?

heycam: yes

anthony: yes, then it is ok.

heycam: I'll check in with test description and pass criteria.

anthony: the test looks fine (after commit).

heycam: test approved?

(no objection)

<scribe> ACTION: heycam to add mask-filter-01 to test suite and implementation report. [recorded in http://www.w3.org/2011/05/05-svg-minutes.html#action05]

<trackbot> Created ACTION-3042 - Add mask-filter-01 to test suite and implementation report. [on Cameron McCormack - due 2011-05-12].

(more discusions between heycam and anthony about the test).

heycam: the last thing on the agenda was the email from Jonathan Watt about SVG 2.0 work and the process.

shepazu: I did not quite fully agree.
... we would need JWatt on the call to discuss this.

<heycam> http://lists.w3.org/Archives/Public/public-svg-wg/2011JanMar/0257.html

shepazu: I suggest we start from a blank slate so that we are aware of what gets put in.
... for SVG 2.0, we need to be very rigorous, and I think that editing the SVG 1.1 spec. is dangerous.
... the work mode needs to match the needs of the editors.

vhardy: what do we need to decide?

shepazu: right now, we need Erik, Chris, Cameron and myself all on the call, at least, to decide about how to start the effort. We also need JWatt on the call

heycam: let's adjourn until we have the right attendance on the call.
... we need to decide this pretty soon.

shepazu: we could start after the AC meeting.

heycam: I'll try to get Jonathan on the call for next week.

shepazu: I also wanted to discuss the scope of the 2.0 effort, as I outlined in email.

<shepazu> http://lists.w3.org/Archives/Public/www-svg/2011May/0007.html

shepazu: I proposed 3 different approaches: 1) integration with CSS, short term, 2) minimal set of features, about 2 years and 3) more significant features, 2 to 3 years.
... I am intersted in the general tenure of what the group thinks and if there is a desire for new features in SVG.
... this is above and beyond what we are doing with the CSS WG (gradients, transforms, compositing).

vhardy: I would suggest we work on the scope, as you suggested, and also turn to the community with a survey.

shepazu: yes, this has been suggested. Does everybody agree that we should set up a poll?

heycam: I agree. I think that would be better than just go with the www-svg feedback?

shepazu: what is the mechanism by which we do this poll? The W3C polls are not really done for that.

anthony: do we need a new polling service to collect this information?

shepazu: one of the things that w3c is doing is adding infrastructure to allow forums.
... people will be able to +1 comments.
... this will take at least a month.

<shepazu> trackbot, end telcon

- DRAFT -

SVG Working Group Teleconference

05 May 2011

Attendees

Contents

F2F venue

Discussion on SVG 2. Based on J Watt's email.

Summary of Action Items

Scribe.perl diagnostic output