SVG Working Group Teleconference -- 16 Nov 2009

<trackbot> Date: 16 November 2009

<anthony> Scribe: anthony

Next SVG F2F

<jwatt> scribe: Jonathan Watt

<jwatt> scribenick: jwatt

AG: we were tentatively thinking 10-14 (Wed-Sun) Feb 2010
... but if that puts too much of a squeeze on Erik's travel, maybe we can move it back a day
... we should start booking soon, because things start to fill up quickly

ED: I would prefer to move it back a day or two
... one day is probably enough
... but I need to check flight times
... worst case, I miss the last day

JW: I'm okay with shifting it back a day

<ed> shepazu: you ok with 9-13th feb?

JW: we should check it meets Doug's constraints too

3D Transforms

http://www.w3.org/mid/200911160854.nAG8sNIH030663@mail16.tpgi.com.au

<anthony> http://lists.w3.org/Archives/Public/www-svg/2009Nov/0025.html

JW: It's not completely clear to me what the second of those two links requires
... but I think it's covered by z-index from what I understand
... waiting on feedback to the email I sent him

<ed> http://www.w3.org/2007/11/SVG_rechartering/SVG-WG-charter.html

<anthony> http://dev.w3.org/SVG/modules/transforms/SVGTransformsReqs.html

JW: do we want to make a formal statement of the extent and limits of SVG's 3D ambitions?

ED: our charter does talk about pseudo 3D

AG: and we have our use cases and requirements document

<ed> "To allow more advanced design features and efficient decorative effects, there will be attention paid to new stroking and filling options, including pseudo-3D (2.5D) effects by means of new gradients, filters, and non-affine transformations, with authoring tips for their use."

AG: that says what we're trying to address

JW: we have the new transforms module and the z-index proposal

ED: and gradients may have some pseudo-3D effects

<scribe> ACTION: anthony to send an email clarifying our pseudo-3D ambitions [recorded in http://www.w3.org/2009/11/16-svg-minutes.html#action01]

<trackbot> Created ACTION-2691 - Send an email clarifying our pseudo-3D ambitions [on Anthony Grasso - due 2009-11-23].

<scribe> ACTION: anthony to update the wiki page with the F2F times [recorded in http://www.w3.org/2009/11/16-svg-minutes.html#action02]

<trackbot> Created ACTION-2692 - Update the wiki page with the F2F times [on Anthony Grasso - due 2009-11-23].

DS: I think we've not done a very good job of articulating our pseudo-3D ambitions
... maybe we need to start using the front page of our wiki to more clearly articulate where we're going

JW: sounds good

DS: I'm going to try to redesign the front page to be much more friendly

<scribe> ACTION: doug to redesign the SVG WG wiki page, and articulate there our pseudo-3D ambitions [recorded in http://www.w3.org/2009/11/16-svg-minutes.html#action03]

<trackbot> Created ACTION-2693 - Redesign the SVG WG wiki page, and articulate there our pseudo-3D ambitions [on Doug Schepers - due 2009-11-23].

DS: we've been using the wiki to collect our thoughts for ourselves, but we should start using it to disseminate information
... in a more friendly form, with links to all the relevant documents
... and we need to have good links from the main page to these pages

JW: agreed

AG: okay, i'll make a page and just link to it in the reply I send to the list

DS: and if Microsoft does join the group some time soon, then probably other companies will join too
... in which case it will be especially useful to have documents that lots of new members can use to catch up more quickly

SVG 1.1 2nd Ed.

ED: I'm going to commit the zip with the Java bindings
... I haven't looked at every single one of them, so I can't guarantee that they are all correct
... I'm planning to email Cameron and Vincent

DS: and ask on the public list for a public review

ED: I can add the diffs
... most of them are fairly small
... but it's still 300 files or so
... I'll do that after the telcon

<ed> http://www.w3.org/Graphics/SVG/WG/wiki/Errata_in_SVG_1.1_Second_Edition

ED: the other thing that needs to be done is review the 3 unreviewed tests on this page
... I haven't had time to do it yet
... if anyone else feels like doing it, feel free
... it may be hard to review that test until we've decided how to resolve the issues in the "Again with the getIntersectionList()" thread on the public list
... so maybe we should schedule that for next week

<ed> http://lists.w3.org/Archives/Public/public-svg-wg/2009OctDec/0005.html

ED: but that's one of the last few things we need to do before we can publish
... maybe a few other's from Cameron's email
... okay, I'll put IntersectionList down for Thursday
... there's nothing major from Cameron's list, I think
... so we should be able to get 2nd Ed out very soon - it would be nice to have it out of the way
... are the harnesses for the test suite in good shape?

AG: yes
... I committed a whole bunch of stuff recently
... all the scripts are up to date
... I removed the svggen directory
... I started going over the Batik render errors
... to make sure the references are getting produced correctly
... I haven't tested the archive generation, but I'm fairly confident it's working correctly

Transforms module feedback

<ed> http://lists.w3.org/Archives/Public/www-svg/2009Nov/0041.html

<anthony> http://dev.w3.org/SVG/modules/transforms/SVGTransforms.html

ED: the first two questions are strongly related to z-index

DS: we should look at how CSS transforms deal with z-index

<ed> http://dev.w3.org/csswg/css3-3d-transforms/#transform-style

<shepazu> [[The position on the Z axis of a transformed element does not affect the order within a stacking context. With elements at the same z-index, objects are drawn in order of increasing z position. ]]

JW: that makes sense to me

DS: and CSS and SVG should be consistent on this
... one way or another

JW: so with transforms one part of the element will appear to be further along the z-axis than another part of the same element

AG: yes

JW: so in principle you could have two elements rotated around the z-axis that could cut through each other

AG: yes, so you have to ask yourself if you want to go full 3D like that
... which means you have to have a completely different rendering model and solve all sorts of problems

JW: you probably need hardware support at that point, which is pobably going further than we want

DS: I don't think we should go to actual 3D
... I think what they've done in CSS transforms is what we should do

<shepazu> [[ ‘transform-style’ property : Note that while ‘preserve-3d’ can be specified on an element, the effect may not be possible. Elements that have ‘overflow’ set to ‘hidden’ are unable to keep their children in 3D. In this case the element will behave as if the property was set to ‘flat’. ]]

<shepazu> [[ This flattening takes place at each element ]]

DS: I think we should articulate this somewhere

JW: the way to think about it is elements are painted one at a time
... as an element is painted, it in by itself given a 3D transform, and painted
... when the next element is painted, the same thing
... this is totally different to taking all elements, arranging them in 3D space, then painting the 3D scene

AG: I think both the CSS spec and our spec should be a lot clearer about this

DS: the SVG and CSS WGs have gotten a lot better at working together
... and I think we need to work a lot closer still with transforms
... so that 3D transforms are completely consistent between the two specs
... possibly one spec would be the way to go
... it would probably be a good place to talk about z-index as well, since how it interacts with 3D transforms needs to be clear

ED: we should mark up our requirements clearly
... it would be interesting to have sections on how things apply to HTML and other sections on how things apply to SVG in the same document

DS: the CSS transforms spec seems stalled for the moment, so making a joint effort on a joint spec could hopefully get things started again
... going back to Steve's email, I think Antony should reply, and put unresolved issues into tracker
... mention that we will be talking to the CSS WG, and that he should also take a look at the CSS WG's spec
... he says he doesn't like matrices, and I don't like them either

AG: we should have them for completeness though

SVG DOM

<shepazu> http://www.w3.org/Graphics/SVG/WG/wiki/DOM

DS: we should add our commentary to the "commentary" page

<shepazu> http://www.w3.org/Graphics/SVG/WG/wiki/DOM_Report

DS: stuff that's good, stuff that should be dropped, stuff to improve...

<scribe> ACTION: anthony to check with JS about joining forces with the CSS WG, and if that's okay, approach them [recorded in http://www.w3.org/2009/11/16-svg-minutes.html#action04]

<trackbot> Created ACTION-2694 - Check with JS about joining forces with the CSS WG, and if that's okay, approach them [on Anthony Grasso - due 2009-11-23].

trackbot: end telcon

- DRAFT -

SVG Working Group Teleconference

16 Nov 2009

Attendees

Contents

Next SVG F2F

3D Transforms

SVG 1.1 2nd Ed.

Transforms module feedback

SVG DOM

Summary of Action Items

Scribe.perl diagnostic output