SVG Working Group Teleconference -- 19 Oct 2010

<trackbot> Date: 19 October 2010

<ed> trackbot, end telcon

<trackbot> Meeting: SVG Working Group Teleconference

<trackbot> Date: 19 October 2010

<scribe> scribeNick: pdengler

Test Suite

ed: Number of check-ins on editors list have come in
... cameron had changes to the spec. What were they?
... There were some styling issues, not content changes

heycam: Erik made a change to the java binding of the interface exceptions were fixed to bind appropriately to java exceptions

<ChrisL> changes I made were for ACTION-2880

<ChrisL> ACTION-2880?

<trackbot> ACTION-2880 -- Chris Lilley to mark all the accepted tests from todays telcon -- due 2010-10-19 -- CLOSED

<trackbot> http://www.w3.org/Graphics/SVG/WG/track/actions/2880

ChrisL: Finished up the action from the last telecon

ed: One comment: One of the filter test already had a patch file

ChrisL: I captured the image from Opera to create the .PNG
... This is an exception and didn't need to happend. Typcially we will hydrate a static SVG file to represent the final image

pdengler: Is this the auto geneated site for tests: http://dev.w3.org/SVG/profiles/1.1F2/test/status/test_suite_status.html

ChrisL: Currently doesn't have all tests, only has the accepted tests

heycam: The script could be modified to include tests not accepted, but it does not at this time

pdengler: We talked last time about cotinuing the evalute the Microsoft sumbitted tests

heycam: I am happy to look at those tests

ChrisL: All that I have looked at I marked as reviewed and we accepted

coords-transformattr-03-f

coords-transformattr-04-f

coords-transformattr-05-f

imp-path-01-f

interact-pevents-10-f

interact-pointer-04-f

linking-a-10-f

masking-mask-02-f

masking-path-12-f

masking-path-13-f

masking-path-14-f

painting-marker-07-f

painting-marker-08-f

painting-marker-properties-01-f

pservers-grad-23-f

pservers-grad-24-f

pservers-grad-stops-01-f

pservers-pattern-04-f

pservers-pattern-05-f

pservers-pattern-06-f

struct-cond-overview-02-f

struct-cond-overview-03-f

struct-cond-overview-04-f

struct-dom-16-f.svg

struct-dom-17-f.svg

struct-dom-18-f.svg

struct-dom-19-f.svg

struct-use-14-f

struct-use-14-f.svg

struct-use-15-f

struct-use-15-f.svg

styling-class-01-f

styling-css-09-f

styling-pres-03-f

styling-pres-04-f

styling-pres-05-f

text-fonts-05-f

types-dom-svgfittoviewbox-01-f

types-dom-svglengthlist-01-f

types-dom-svgnumberlist-01-f

types-dom-svgstringlist-01-f

types-dom-svgtransformable-01-f

pdengler: Above are the tests I believe that haven't yet been reviewed that Microsoft submitted

<ChrisL> or as an alternative, grep Microsoft *.svg | grep -v approved

TPAC

ed: Who will be at TPAC

pdengler, ed, schepazu, anthony, chris

shepazu: We should schedule a tuesday meetng with CSS

ChrisL: I discussed this with the CSS telecon; we want to have a meeting, but haven't finalized that

ed: I cannot make Tuesday

shepazu: Maybe we can allocate our time to meet with them instead
... Chris, can you go back and say we want to meet on Thursday

ChrisL: I will propose that as a date

I will submit a high level proposal on a scoped approach to transforms, transitions, animations and the SVG DOM which should complement all efforts to date

ChrisL: Fonts doesn't need anything specific in the spec that would affect SVG. WOFF support should be just mentioned as required in SVG 2.0

ed: Does your white paper cover 3d transforms

pdengler: No not now, it's a scoping issue

shepazu: If your paper doesn't cover Filters, paint servers, we should possibly discuss that as well
... We shoud look at these with CSS working group

http://www.w3.org/Graphics/fx/

http://www.w3.org/Graphics/fx/charter/

transform coordinates

shepazu: Currently getCoordsAt api was reivewed with DOML3 event WG
... The other folks didn't agree it was in scope
... Failing that, the most natural place to put this method is in the transforms spec

http://dev.w3.org/Graphics-FX/modules/2D-transforms/spec/2DTransforms.html

shepazu: Original thought would be that there would be a couple of attributes to support the transformed coordinate space (absX,absY)
... on the mouse event. And then return you an X and Y
... jwatt had a proposal by using a method (getCoordsAt)

pdengler: can't we just share the code that jwatt provided that unwinds the transforms

anthony: We should but that in the spec.

pdengler: And would that apply to nested content HTML/SVG/HTML/SVG

shepazu: Yes, it would probably be on Element, which inherits from node

anthony: now is the time to do it

<ChrisL> /me fixed the encoding of the fx charter btw

<scribe> ACTION: anthony to add getCoordsAt to transform spec and coordinate with shepazu [recorded in http://www.w3.org/2010/10/19-svg-minutes.html#action01]

<trackbot> Created ACTION-2883 - Add getCoordsAt to transform spec and coordinate with shepazu [on Anthony Grasso - due 2010-10-26].

href , xlink:href

<shepazu> http://lists.w3.org/Archives/Public/public-svg-wg/2010OctDec/0029.html

pdengler: We do not want to change the casing of the HTML parser and what it serialized out as

<ChrisL> ok so in xml, href is the only spelling that is accepted. for html serialisation, HrEf will get serialized as href

shepazu: This actually touches on something else. If I didn't want to have xlink: as my namespace
... hard coded prefixes is a reasonable fix to this problem

<ChrisL> rdf uses a tone of namespaces

shepazu: There are a handful of common prefixes
... The one I care about is xlink.

<adrianba> HTML5 defines some attributes with a fixed prefix: http://dev.w3.org/html5/spec/syntax.html#attributes-0

<shepazu> http://www.w3.org/TR/html5/namespaces.html#svg-namespace

Animations and baseValues

shepazu: I would like to start doing something on animation in the fx task force

heycam: Me too.

shepazu: are we going to use the 'lacuna' approach in SVG 2.0
... e.g. If you don't specify width, and you animate width, what does it start with. That's not said in the spec

ed: I think it is set in the spec

<ed> implied by the animVal DOM

shepazu: I used to think that the SMIL worked on the same presentation model as CSS, but it does not
... now that we are going to specify it, are we going to have two different models for the two different syntaxes, or could we say that they are one and the same.
... That means we only have one way to get the presentation value
... SMIL doesn't care what the OM is.
... If we have the freedom, we could move forward with using CSS as the animation underneath the DOM
... What if we exposed our attributes through the CSS OM for purposes of animation

ChrisL: If you can find some good wording that represents that

Test Status (with Anthony on the call now0

anthony: I commited about 10 tests a minute ago

<scribe> ACTION: anthony to modify the script to allow different test suites status to be created [recorded in http://www.w3.org/2010/10/19-svg-minutes.html#action02]

<trackbot> Created ACTION-2884 - Modify the script to allow different test suites status to be created [on Anthony Grasso - due 2010-10-26].

<shepazu> http://tech.groups.yahoo.com/group/svg-developers/message/64113

<ed> http://dev.w3.org/SVG/profiles/1.1F2/test/svg/struct-image-16-f.svg

<heycam> http://mcc.id.au/temp/ser.html -- a test to see how xlink:href is serialized in html

<ed> http://dev.w3.org/SVG/profiles/1.1F2/test/svg/struct-image-05-b.svg

<ChrisL> coords-transformattr-05-f.svg

<ChrisL> I found five tests with grep \<image *.svg | grep .svg\"

<ChrisL> a test that points to an image with script, and the pass is that the script does not run

<ed> trackbot, end telcon

- DRAFT -

SVG Working Group Teleconference

19 Oct 2010

Attendees

Contents