See also: IRC log
<trackbot> Date: 20 July 2010
<shepazu> minutes
<ed> Agenda: http://lists.w3.org/Archives/Public/public-svg-wg/2010JulSep/0047.html + http://lists.w3.org/Archives/Public/public-svg-wg/2010JulSep/0048.html
<ed> http://www.w3.org/Graphics/SVG/WG/wiki/Paris-2010-F2F
<ed> http://www.w3.org/2002/09/wbs/19480/f2fParis2010/
<ed> ACTION: ed to fix broken images in 1.1F2/images (commit with -kb flag) [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action01]
<trackbot> Created ACTION-2825 - Fix broken images in 1.1F2/images (commit with -kb flag) [on Erik Dahlström - due 2010-07-27].
<ed> http://lists.w3.org/Archives/Public/public-svg-wg/2010JulSep/0042.html
<ed> http://lists.w3.org/Archives/Public/public-svg-wg/2010JulSep/0043.html
<ed> http://lists.w3.org/Archives/Public/public-svg-wg/2010JulSep/0045.html
<scribe> scribenick: pdengler
pdengler: CSS defined properties
(e.g. 'font-size') must follow CSS syntax and thus require
units for non-zero <length> values. See 6.1 for a list of
CSS defined properties
... This is the proposed change of language correct?
ed: Not a problem going to a property for x and y
shepazu: When using attributes use SVG unit values, when using CSS use unit values
ed: not sure this is what browers do
Tav: Not sure if this is what we are saying
<ed> http://dev.w3.org/SVG/profiles/1.1F2/test/svg/types-basic-02-f.svg
shepazu: Need to coordinate w/
CSS working group
... Take HTML approach that CSS working group needs to reasses
the issue
... Need to do with the Fx task force
... Tooling and existing content does styling with unitless
values.
<scribe> ACTION: shezpazu to Work with Tav and CSS working group on understanding the requirment for CSS/SVG Unitless Properties [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action02]
<trackbot> Sorry, couldn't find user - shezpazu
<scribe> ACTION: shepazu to Work with Tav and CSS working group on understanding the requirment for CSS/SVG Unitless Properties [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action03]
<trackbot> Created ACTION-2826 - Work with Tav and CSS working group on understanding the requirment for CSS/SVG Unitless Properties [on Doug Schepers - due 2010-07-27].
ed: Any other last call items; perhaps one or twomore
<ed> ISSUE-2353
ed: Wasn't clear from the description what the red-stroked rectangles were suppose to describe
<scribe> ACTION: to ed Make modification to spec around feImage coordinates regarding issue 2353 [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action04]
<trackbot> Sorry, couldn't find user - to
<scribe> ACTION: ed To make modification to spec around feImage coordinates regarding issue 2353 [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action05]
<trackbot> Created ACTION-2827 - Make modification to spec around feImage coordinates regarding issue 2353 [on Erik Dahlström - due 2010-07-27].
ed: With no more last call items
for today we will move on to other issues
... Topic : Default sizing of SVG in HTML5
shepazu: If default size is 100% 100%, if there is a view box, you can infer size
ed: not necessarily size
... viewbox infers ratio, not size
... It would help to have an example of what you expect and
what you get
... We should have samples in the test suite
pdengler: They are there, but they are not necessarily complicated enough
<scribe> ACTION: pdengler to create more sophisticated test cases around <image> sizing and viewBox [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action06]
<trackbot> Created ACTION-2828 - Create more sophisticated test cases around <image> sizing and viewBox [on Patrick Dengler - due 2010-07-27].
pdengler: When you have a simple <svg><circle...> the default size is 300px,150px as last fall back
ed: Will resolve differently depending on outer container
shepazu: Need to have default intrinsic height in 2.0
<scribe> ACTION: patd to investigate default <svg> sizing in HTML5 [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action07]
<trackbot> Sorry, couldn't find user - patd
<scribe> ACTION: pdengler to investigate default <svg> sizing in HTML5 [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action08]
<trackbot> Created ACTION-2829 - Investigate default <svg> sizing in HTML5 [on Patrick Dengler - due 2010-07-27].
shepazu: For canvas, since SVG is a "vector", I think the drawImage should take a width and height parameter
ed: It does take a width and height, but it doesn't always have them available
shepazu: In the case where there isn't a width and height, we need to determine what to do
<ed> http://lists.w3.org/Archives/Public/www-svg/2010Jul/0074.html
shepazu: If it has a width and height, use the width and height of the viewBox in pixels
ed: Not sure this is a great
idea
... If you don't pass a width and height when you call
drawImage, the SVG image needs to be drawn into an area, you
could use the <canvas> width and height
... Then the underlying viewBox could still be used for the
coordinate system
... We have experimental code dealing with this which probably
needs further fixing
pdengler: I like this.
ed: We should get this back to the HTML/Canvas effort
<scribe> ACTION: pdengler to get to HTML5 working group <canvas> about using canvas width and height for SVG width and height [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action09]
<trackbot> Created ACTION-2830 - Get to HTML5 working group <canvas> about using canvas width and height for SVG width and height [on Patrick Dengler - due 2010-07-27].
ed: We need to get jwatt's opinion as well
pdengler: I will use this as a first poposal and be prepared to back it out of Jonothan does not like it
<ed> http://lists.w3.org/Archives/Public/www-svg/2010Jul/0070.html
<ed> ISSUE-2254?
<trackbot> ISSUE-2254 -- Consider changing which elements have load dispatched to them -- raised
<trackbot> http://www.w3.org/Graphics/SVG/WG/track/issues/2254
ed: What we concluded that
loadevents on elements that have external resources; for other
elements we couldn't care less; but that's not what the spec
says
... And there is at least one test that checks for these
<ed> http://lists.w3.org/Archives/Public/www-svg/2010Jul/0073.html
<scribe> ACTION: to pdengler Check into SVG Load on all elements and performance [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action10]
<trackbot> Sorry, couldn't find user - to
<scribe> ACTION:pdengler to Check into SVG Load on all elements and performance [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action11]
<trackbot> Sorry, bad ACTION syntax
<scribe> ACTION:pdengler to Check into SVG Load on all elements and performance [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action12]
<trackbot> Sorry, bad ACTION syntax
<scribe> ACTION: pdengler to Check into SVG Load on all elements and performance [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action13]
<trackbot> Created ACTION-2831 - Check into SVG Load on all elements and performance [on Patrick Dengler - due 2010-07-27].
<scribe> Topic : Markers and styles at point of reference
<ed> http://lists.w3.org/Archives/Public/www-svg/2010Jul/0066.html
pdengler: We should be using markers styled at the point of reference
shepazu: There is an issue that
there is a lot of content out there that already follows
this
... The way we talked about it, was to have a different method
for styling the markers
<ed> http://dev.w3.org/SVG/profiles/1.1F2/test/svg/painting-marker-02-f.svg
pdengler: Looking at this test http://dev.w3.org/SVG/profiles/1.1F2/test/harness/htmlEmbed/painting-marker-02-f.html
<ed> <marker id=foo><rect fill="currentColor"></marker> <line marker-start="url(#foo)" color="blue"/>
shepazu: unforutnately, the
conceptual model that a marker was not a <use>
... Tried a number of ways to change fill/stroke of
markers
... As a forward change in SVG 2.0, but we could add this in
one of two ways;
... Keep it backward compatible with marker-fill/marker-stroke;
this would support both use cases
... Another way to do it is to inherit the fill and stroke of
it's referencing parent
Tav: You can't just inherit the stroke of the parent
shepazu: most content defines the
style for the marker.
... you don't get to override this. It's not as much as an
issue as I thought, as existing content is already defining
stroke/stroke-width colors
<ed> ok, we can forget the marker + color example i pasted earlier, doesn't work :)
shepazu: So we could redefine it for SVG 2.0; but for the author is best for us define marker specifc properties
<scribe> ACTION: ed: to investigate markers and where and when to apply style [recorded in http://www.w3.org/2010/07/20-svg-minutes.html#action14]
<trackbot> Created ACTION-2832 - Investigate markers and where and when to apply style [on Erik Dahlström - due 2010-07-27].
Tav: Suppose there is an arrow. The normal thing to do would be to have the arrow head to inherit the stroke-color of the path for the fill of the marker
shepazu: How do we achieve that
effect?
... Most of the time, authors are using the path without a
fill
This is scribe.perl Revision: 1.135 of Date: 2009/03/02 03:52:20 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Found ScribeNick: pdengler Inferring Scribes: pdengler Default Present: ed, +33.9.53.77.aaaa, anthony, +1.919.824.aabb, tav, shepazu, [Microsoft] Present: ed +33.9.53.77.aaaa anthony +1.919.824.aabb tav shepazu [Microsoft] Regrets: CL WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 20 Jul 2010 Guessing minutes URL: http://www.w3.org/2010/07/20-svg-minutes.html People with action items: ed make modification patd pdengler shepazu shezpazu to[End of scribe.perl diagnostic output]