W3C home > Mailing lists > Public > www-svg@w3.org > August 2012

Minutes, August 23 2012 telcon

From: Rik Cabanier <cabanier@adobe.com>
Date: Thu, 23 Aug 2012 15:31:09 -0700
To: "www-svg@w3.org" <www-svg@w3.org>
Message-ID: <83F37C1A4497B54589EAEDC750D03A9473B7E74E@nambx09.corp.adobe.com>
Here are the meeting minutes of the August 23 2012 telcon:


      [1] http://www.w3.org/

                               - DRAFT -

                    SVG Working Group Teleconference

23 Aug 2012


      [2] http://lists.w3.org/Archives/Public/public-svg-wg/2012JulSep/0167.html

   See also: [3]IRC log

      [3] http://www.w3.org/2012/08/23-svg-irc


          Doug_Schepers, heycam, Kelvin, Rich, cabanier, Tav, ed,
          nikos, birtles




     * [4]Topics
         1. [5]SVG2 status
         2. [6]Lucerne F2F reminder
         3. [7]SVG Accessibility Gap Analysis
         4. [8]CSS4 <image> as value for <paint>
         5. [9]<color-profile> element
     * [10]Summary of Action Items

   <trackbot> Date: 23 August 2012

   <ed> Agenda:

     [11] http://lists.w3.org/Archives/Public/public-svg-wg/2012JulSep/0167.html

   <scribe> scribenick: cabanier

SVG2 status

   heycam: we were planning on publishing last tuesday but there
   were some issue
   ... we tried again on thursday but we had some issues with
   Chris' color changes. I waiting to hear back from him and we'll
   probably publish on tuesday
   ... because we have inline svg and we have mathml
   ... we needed to change the format of the doc to HTML5 which
   took a bit of time

   cabanier: so we're using mathml?

   heycam: I converted a couple of formulas of the painting
   formulas and some of of the path stuff
   ... I'm using mathjax
   ... to provide fallback.
   ... and around 6 months ago it was decided that mathml was OK
   ... are you having trouble?

   tav: I tried it today and it took me around 20 min

   heycam: let me know if you have issues and I'll help you

Lucerne F2F reminder


     [12] http://www.w3.org/Graphics/SVG/WG/wiki/F2F/Switzerland_2012


     [13] http://www.w3.org/Graphics/SVG/WG/wiki/F2F/Switzerland_2012/Agenda_proposals

   ed: there's a page for attending and for proposals

   heycam: I think only cyril hasn't replied

SVG Accessibility Gap Analysis


     [14] http://www.w3.org/mid/OF38278C69.E13ACB0C-ON86257A5E.006A2B07-86257A5E.006B7B8B@us.ibm.com

   <richardschwerdtfe> Rich

   heycam: can I reformat it as text and send it to the list

   rich: sure

   doug: that will make it easier to search the archives


     [15] http://lists.w3.org/Archives/Public/www-svg/2012Aug/0105.html

   rich: the first part is just about the technology of what
   happens in the browser
   ... looking at SVG, we have the gap analysis

   <shepazu> (it might be useful to also look at this
   [16]http://www.w3.org/TR/2011/WD-html-aapi-20110414/ )

     [16] http://www.w3.org/TR/2011/WD-html-aapi-20110414/

   rich: the reason for the role attribute is to give the system

   (rich: going over the reason we need accessibility in SVG)

   rich: we have no keyboard navigation
   ... the structure of SVG doesn't map to platform accessibility
   ... the grouping element can be repurposed for that reason
   ... you can assign semantics to it
   ... the first svg document alludes that we can do this with the
   <g> element
   ... svg rendering don't follow system settings such as high
   ... does high contrast tie in with the developers?

   <heycam> I wonder whether the CSS system colour keywords in
   conjunction with default colours is sufficient for high
   constrast support.

   rich: panning and zooming inside of a SVG rendering is
   important too
   ... do we do something with animations
   ... we heard that they are being removed?

   heycam: no. Brian is working on something new

   rich: we want to look into how to make these accessibility
   ... we should introduce 'role' on elements
   ... the role attribute on safari and webkit on svg is mapped to
   the accessibility API
   ... we could develop more states to ARIA and apply them to SVG
   ... MS and ??? would like tabindex applied to SVG so you can
   have keyboard navigation
   ... do we want to provide the ability to zoom into content that
   is not visible

   doug: what is the behavior of fragment id with zooming and
   ... the user might not want to zoom in on the bounding box.
   ... we've talked about the idea of adding tab index to SVG but
   SVG already has one that is better

   <ed> an example of tabbing between nodes in an svg, with

     [17] http://dev.w3.org/SVG/profiles/1.2T/test/svg/interact-focus-203-t.svg

   <ed> works in opera

   rich: so you want both models?

   doug: yeah

   <ed> (testcase is from the SVG 1.2T testsuite)

   rich: that sounds fine. I have to look into again
   ... but there doesn't seem opposition to adding tab index

   heycam: yes, we should have keyboard navigation. Even outside
   the accessibility use case

   ed: do we just want tab behavior or 4/8 way navigation too?
   ... I'd prefer that whatever we had in tiny keeps working

   heycam: we should look into the model

   ed: yes, it should be possible to make it work. It makes sense
   to have it in SVG if it works in HTML

   heycam: weren't they going to be added to CSS?

   doug: tantek decided not to add the navigation keyword to CSS
   UI 3 but I don't know why
   ... tab index is not sufficient for a lot of scenarios. We
   would have to define how those work together
   ... for arrow key navigation, the keyword are: up/right/etc
   ... those are directional navigation elements

   mmitments -- see items 97, 98, 99 -- that covers the navigation
   properties and HTML's tabindex

     [18] http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Requirements_Commitments

   doug: you need an algorihtmic way to find out what 'up' means

   heycam: item 99 is similar to tabindex, 97 covers nav

   rich: so it's resolved to do this feature?

   heycam: yes

   rich: who is going to do the spec work?

   heycam: nobody has signed up for it


     [19] http://www.w3.org/TR/SVGTiny12/interact.html#specifyingnavigation

   doug: here's what we have so far
   ... the only conflict is nav-next and nav-prev
   ... I will take a look if there's a conflict with tab index.
   It's an interest of mine

   <scribe> ACTION: shepazu is going to look into conflicts
   between the model of tab index and navigation [recorded in

   <trackbot> Created ACTION-3341 - Is going to look into
   conflicts between the model of tab index and navigation [on
   Doug Schepers - due 2012-08-30].

   doug: I see they're attributes but not properties. We need to
   talk to the CSS people about that

   rich: I''m willing to take an action for the aria and the role

   heycam: we already have role in tiny


     [21] http://www.w3.org/TR/SVGTiny12/struct.html#RoleAttribute

   tav: please provide examples because I tried to do it in the
   past and there's no documentation out there

   doug: I have some!

   tav: those are the only ones...

   doug: I dropped in a document. This contains a table that has
   all the ally api's
   ... I found it really useful

   rich: yes, everyone needs to understand what we're doing

   resolution: define Role and ARIA states and property attributes
   in SVG

   doug: we want to do the same as HTML did it

   rich: we feel that the industry hasn't done enough for
   ... with flash going away, we want to look into SVG

   heycam: what did flash do?

   rich: it doesn't deal with table structure. You can have a
   group or a panel. The evens are hardcoded to the windows

   <shepazu> example of SVG table with ARIA

     [22] http://schepers.cc/svg/accessibility/table.svg

   rich: it has no knowledge about rich text.
   ... they don't have structure content
   ... it was not that great and Adobe is not investigating it any
   ... limited windows support. They wanted to add richer
   accessibility support but it's not happening

   <Zakim> shepazu, you wanted to discuss system colors

   doug: what did you mean with system colors?

   rich: desktop applications can respond to going to high
   contrast mode

   <shepazu> CSS 3 system colors deprecated

     [23] http://www.w3.org/TR/css3-color/#css-system

   rich: we could add an API for that so the application is

   doug: I think there's an opportunity for doing this in CSS
   ... like with a media query
   ... not just for ally, but various lighting conditions
   ... or for people that are color blind
   ... using media queries we can have the document be styled

   <birtles> CSS luminance media queries
   SS-Media-Queries-971/ (doesn't support the red-green red-blue
   thing though)

     [24] http://fremycompany.com/BG/2012/Ambient-light-sensors-and-CSS-Media-Queries-971/

   doug: then designers can provide an alternate representation

   tav: do you think designers will do this?

   doug: I think there will be aftermarket people will do this
   ... designers don't want to do the high contrast page because
   it doesn't look good

   ed: I like Brian's doc

   rich: we should go to the CSS WG for the media queries?

   all: yes

   <scribe> ACTION: richardschwerdtfe to write up a proposal for
   role and aria states and properties integration [recorded in

   <trackbot> Created ACTION-3342 - Write up a proposal for role
   and aria states and properties integration [on Richard
   Schwerdtfeger - due 2012-08-30].

   rich: panning and zooming is something that Kelvin wants
   ... it would be nice to pan and zoom portions
   ... of an SVG document

   kelvin: you could describe a rectangle to do pan and zoom
   ... it's not an SVG thing. I'm not sure how the WG can
   influence this. The Adobe viewer had

   ed: it's supported in some other viewers. Not sure if it's well
   documented in the respective UA's docs how it works

   rich: can we put this in the spec?

   heycam: it's still in the spec but it hasn't gotten widespread

   doug: in the past people wanted to limit the scope that we can
   influence the UI

   <ed> s/it's supported in some other viewers/ctrl+dragging
   inside an svg for panning, alt+click / shift+alt+click for
   zooming in and out is supported in some other viewers/

   doug: Opera has the zoom slider
   ... we can try to say that they need to implement some sort of
   zoom and pan

   kelvin: with accessibility we can say that we need it which
   might force them to add it

   rich: should we strengthen the language?

   doug: Kelvin, would you want to write up what we need from pan
  and zoom? so we can feedback from browser vendors

   kelvin: yes, I will reread the spec and write something up

   <scribe> ACTION: kelvin to write requirement for zoom and pan
   for accessibility [recorded in

   <trackbot> Created ACTION-3343 - Write requirement for zoom and
   pan for accessibility [on Kelvin Lawrence - due 2012-08-30].

CSS4 <image> as value for <paint>


     [27] http://www.w3.org/mid/FBBDBDBC-F3CD-404E-AFC1-0369A75DAA89@adobe.com

   ed: why was this for css4 images and not 3?

   cabanier: isn't 3 already in LC? Tab is working on version 4

   heycam: so, the element() is in CSS4

   cabanier: maybe put it on the agenda again next week

   <ed> ed: css3-images is in CR

   cabanier: wasn't there also a decision to push off element()
   ... maybe it was the one that referred to other html chunk

   ed: we
   ... we'll put it on the agenda for next week

<color-profile> element

   ed: chris isn't here

   heycam: we already decided to remove the font element
   ... there is no advantage to be able to specify as an element
   as well as a CSS @ rule

   cabanier: that sounds reasonable

   heycam: and it wasn't widely implemented

   ed: is this tagging the image with a profile?


     [28] https://svgwg.org/svg2-draft/color.html#ColorProfileElement

   cabanier: it's there so you can give it a link to an ICC
   profile and give it a name

   heycam: and then you can reference the name of the color
   profile in the paint
   ... it makes sense just to have the @ rule

   ed: I can't think of a reason to have it

   heycam: it's just as easy to get to an @ rule through the DOM

   resolution: remove the <color-profile> element from the spec

   <scribe> ACTION: heycam to remove <color-profile> element from
   the spec [recorded in

   <trackbot> Created ACTION-3344 - Remove <color-profile> element
   from the spec [on Cameron McCormack - due 2012-08-30].

   ed: any other topics?

   heycam: there are some things on the mailing list but we can do
   that later

   nikos: weren't there things that we want Erik's input on?

   heycam: one of the items was clipping on filters:

   nikos: we didn't decide

   ed: we discussed it a lot. the current version of the spec
   picks one of the ways and we discussed doing auto for drop
   shadows etc

   heycam: we have margins too.

   ed: we took them out but re-added them for shaders
   ... I have to look into it again

   heycam: dirk should be here to drive the discussion


     [30] http://www.w3.org/2012/08/09-svg-minutes.html#item07

   heycam: this is the discussion

   <ed> ACTION: ed to respond to the filter issues raised in
   [recorded in

     [31] http://www.w3.org/2012/08/09-svg-minutes.html#item07

   <trackbot> Created ACTION-3345 - Respond to the filter issues
   raised in
   [33]http://www.w3.org/2012/08/09-svg-minutes.html#item07 on
   Erik Dahlström - due 2012-08-30].

     [33] http://www.w3.org/2012/08/09-svg-minutes.html#item07


     [34] http://www.w3.org/2012/08/16-svg-minutes.html#item04

   ed: I will read these offline and respond

   <ed> trackbot, end telcon

Summary of Action Items

   [NEW] ACTION: ed to respond to the filter issues raised in
   [recorded in
   [NEW] ACTION: heycam to remove <color-profile> element from the
   spec [recorded in
   [NEW] ACTION: kelvin to write requirement for zoom and pan for
   accessibility [recorded in
   [NEW] ACTION: richardschwerdtfe to write up a proposal for role
   and aria states and properties integration [recorded in
   [NEW] ACTION: shepazu is going to look into conflicts between
   the model of tab index and navigation [recorded in

     [35] http://www.w3.org/2012/08/09-svg-minutes.html#item07

   [End of minutes]

    Minutes formatted by David Booth's [41]scribe.perl version
    1.136 ([42]CVS log)
    $Date: 2012/08/23 22:27:10 $

     [41] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
     [42] http://dev.w3.org/cvsweb/2002/scribe/

Scribe.perl diagnostic output

   [Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.136  of Date: 2011/05/12 12:01:43
Check for newer version at [43]http://dev.w3.org/cvsweb/~checkout~/2002/

     [43] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/rule/role/
Succeeded: s/ well documented/ well documented in the respective UA's do
cs how it works/
WARNING: Bad s/// command: s/it's supported in some other viewers/ctrl+d
ragging inside an svg for panning, alt+click / shift+alt+click for zoomi
ng in and out is supported in some other viewers/
Succeeded: s/the topic next/the agenda for next/
Succeeded: s/I will respond to both offline/I will read these offline an
d respond/
Found ScribeNick: cabanier
Inferring Scribes: cabanier
Default Present: Doug_Schepers, heycam, Kelvin, Rich, cabanier, Tav, ed,
nikos, birtles
Present: Doug_Schepers heycam Kelvin Rich cabanier Tav ed nikos birtles
Agenda: [44]http://lists.w3.org/Archives/Public/public-svg-wg/2012JulSep

     [44] http://lists.w3.org/Archives/Public/public-svg-wg/2012JulSep/0167.html

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 23 Aug 2012
Guessing minutes URL: [45]http://www.w3.org/2012/08/23-svg-minutes.html
People with action items: ed going heycam is kelvin richardschwerdtfe sh

     [45] http://www.w3.org/2012/08/23-svg-minutes.html

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.

   End of [46]scribe.perl diagnostic output]

     [46] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
Received on Thursday, 23 August 2012 22:31:47 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 8 March 2017 09:47:29 UTC