W3C

- DRAFT -

SVG Working Group Teleconference

22 May 2014

Agenda

See also: IRC log

Attendees

Present
cabanier, krit, Tav, ed, stakagi, nikos, Rich
Regrets
Chair
ed
Scribe
ed, nikos, krit

Contents


<trackbot> Date: 22 May 2014

<krit> need a minute

<ed> scribeNick: ed

Cleaning up SVG2

http://lists.w3.org/Archives/Public/www-svg/2014May/0022.html

krit: we discussed on ML, want to remove sections...
... first part ...(missed)
... second part, we should make css always convert to svg (???)

<nikos> scribenick: nikos

Tav: 2 issues here
... you want all css stuff in one place. I don't think that's a good idea
... text stuff should be in the text section
... I want to see what css modules apply in the text section, not hidden elsewhere in the spec
... second issue is that if someone is looking at the spec, it's good to see what the css modules are being used for
... e.g. filter chapter should have small section with explanation of what filter is and one or two examples and then refer to css spec
... this way someone looking at the chapter can get a quick idea of what it's about
... you also asked that text sections be moved. e.g. text-decoration
... text-decoration has problems interacting with svg
... would be great if that's solved in the spec but I still think we should have examples
... what parts of css modules apply to what versions of svg?
... to answer that we have to have something in the svg spec stating what parts of css are used
... I don't think you can just remove stuff that is defined by svg
... I think text-indent could be removed, but there's details to be discussed. does it apply to svg 1.1 style text or just when you have a region for layout?

krit: I don't want to get into detail for each section
... in general I don't think we should preserve svg 1.1. style
... we should remove anything that is duplicated
... e.g. filters has no content
... all the sections with no content can be removed in my eyes
... of course there are links that need to be defined
... how certain things apply to svg needs to be defined
... but it doesn't make sense to have text for everything
... going back to filter effects example, I think examples can be in the filter effects spec
... it's not a handbook, it's a specification for implementers
... even examples can be out-dated
... why not just say you can apply filter effects and link to the spec
... we have nothing about blending in the svg spec yet blending applies to svg

Tav: I think we should have short sections for each. I hate having to click through multiple specifications

krit: that's always going to be the case, whether examples are given or not
... we need an interactions module that explains how parts of other specs are used
... we need normative text if there is svg specific behaviour

Tav: I think when you look at what can be cut it will be quite limited

krit: well I definitely don't want to cut everything

Tav: when I look through the css specs, fonts 3 is a good example, it has lots of examples and I find it easy to understand

krit: every other css spec should be the same
... svg should concentrate on svg itself
... we agree we need a section that explains the interaction with the css modules

Tav: yes

krit: can we agree sections with no content but just links to other specs. e.g. text-align

Tav: I agree that when the spec is done we can clean that up. but dont agree that we can remove right away
... we need to look at each section as a whole and work out what is valuable to keep

<krit> scribenick: krit

krit: Can we agree that each section needs a review as well as the CSS counter part and that we need to check the whole CSS module if and how it applies to SVG before we remove anything?

Tav: Yes, absolutely

<scribe> ACTION: Tav to review CSS Text decoration and CSS3 Text and compare it with SVG [recorded in http://www.w3.org/2014/05/22-svg-minutes.html#action01]

<trackbot> Created ACTION-3626 - Review css text decoration and css3 text and compare it with svg [on Tavmjong Bah - due 2014-05-29].

<scribe> ACTION: krit to review CSS Text decoration and CSS3 Text and compare it with SVG [recorded in http://www.w3.org/2014/05/22-svg-minutes.html#action02]

<trackbot> Created ACTION-3627 - Review css text decoration and css3 text and compare it with svg [on Dirk Schulze - due 2014-05-29].

text-decoration-color

Tav: problem is that the property allows a single color to be defined to a text decoration
... SVG on the other hand allows fill and stroke for the text decoration
... how could we make these apply with text-decoration-color
... it would desireable to define a pattern or gradient on the text decoration

ed: so the fill of text on the stroke of text decoration?

Tav: something like that

<ed> I meant: text-decoration-fill and text-decoration-stroke

krit: don't we have context-stroke and context fill on svg fonts?

Tav: yes
... but a values

krit: right, could text-decor-color take these keywords?
... you can not define stroke and fill separately for decoration

Tav: but does it apply to the stroke or fill of the text decoration itself
... it would be interesting to know if fill and stroke will apply to text in cSS as well

krit: 1) yes fill and stroke will apply to html text in the future

Tav: in a way compatible to SVG

krit: we should make sure that this is the case for the CSS WG

Tav: concerned that this will not be the case

krit: filla nd stroke will apply to text in CSS as it applies to SVG elements today
... would the SVG WG accept to change initial value for fill to currentColor
... that could solve the problem with color and fill on text for HTML and SVG

ed: what is the value of color if not defined?

krit: it is an inheriting property
... ancestor sets color
... it no ancestor does, it is black

<ed> scribeNick: ed

ed: are the benefits worth it? what problems would it solve?

nikos: I think dirk meant that you could then specify the fill and stroke separately
... currentColor seems a bit weird

<nikos> nikos: I think Dirk was meaning that you could set text-decoration-fill without setting text-decoration-stroke and if stroke was enabled then it would get a sensible value from somewhere

ed: another question is: how important is it to be able to control fill and stroke separately for text-decoration

tav: you can already do it using another tspan
... simple thing to fix if we have CSSWG on board

ed: so this is option 3 in your writeup, correct?

tav: yes, combined with option 2, setting 'text-decoration-color' set text 'fill'

ed: ok, let's see what feedback the css editors give us on that proposal

textLength and <tspan>

http://lists.w3.org/Archives/Public/www-svg/2014May/0011.html

ed: there's variety of behavior for textLength combined with tspan, I'm thinking that the spec might be too vague
... tav what do you think?
... it gets a little tricky when you have textLength both on parent <text> and on a child <tspan>
... not sure the spec covers that in enough detail

tav: you'd use the textlenght from the tspan, and then use that to go back to the parent

nikos: yes, that's the way i read it too

ed: we have no tests for this in the testsuite yet

tav: suggest that thomas proposes some clarified spec text, and a test

<scribe> ACTION: thomas to propose clarifications for textLength and <tspan>, and to submit some tests to the svg2 testsuite for this [recorded in http://www.w3.org/2014/05/22-svg-minutes.html#action03]

<trackbot> Created ACTION-3628 - Propose clarifications for textlength and <tspan>, and to submit some tests to the svg2 testsuite for this [on Thomas Smailus - due 2014-05-29].

vertical-align, baseline-shift, alignment-baseline, dominant-baseline

tav: inkscape has baseline-shift for sub and super scripts

ed: what about alignment-baseline / dominant-baseline?

tav: don't think so

ed: CL sent a mail with the conclusion from csswg on letting svg2 define these properties for now

tav: fine with me

trackbot, end telcon

Summary of Action Items

[NEW] ACTION: krit to review CSS Text decoration and CSS3 Text and compare it with SVG [recorded in http://www.w3.org/2014/05/22-svg-minutes.html#action02]
[NEW] ACTION: Tav to review CSS Text decoration and CSS3 Text and compare it with SVG [recorded in http://www.w3.org/2014/05/22-svg-minutes.html#action01]
[NEW] ACTION: thomas to propose clarifications for textLength and <tspan>, and to submit some tests to the svg2 testsuite for this [recorded in http://www.w3.org/2014/05/22-svg-minutes.html#action03]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2014/05/22 13:56:38 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.138  of Date: 2013-04-25 13:59:11  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/krit/Tav/
Succeeded: s/text in future/html text in the future/
Found ScribeNick: ed
Found ScribeNick: nikos
Found ScribeNick: krit
Found ScribeNick: ed
Inferring Scribes: ed, nikos, krit
Scribes: ed, nikos, krit
ScribeNicks: ed, nikos, krit
Default Present: cabanier, krit, Tav, ed, stakagi, nikos, Rich
Present: cabanier krit Tav ed stakagi nikos Rich
Agenda: http://lists.w3.org/Archives/Public/www-svg/2014May/0039.html
Found Date: 22 May 2014
Guessing minutes URL: http://www.w3.org/2014/05/22-svg-minutes.html
People with action items: krit tav thomas

[End of scribe.perl diagnostic output]