W3C home > Mailing lists > Public > public-svg-wg@w3.org > January to March 2012

Minutes, 16 Feb 2012 SVG telcon

From: Chris Lilley <chris@w3.org>
Date: Thu, 16 Feb 2012 22:33:54 +0100
Message-ID: <1427243666.20120216223354@w3.org>
To: public-svg-wg@w3.org
Hello ,


                   SVG Working Group Teleconference

16 Feb 2012


      [2] http://lists.w3.org/Archives/Public/public-svg-wg/2012JanMar/0071.html

   See also: [3]IRC log

      [3] http://www.w3.org/2012/02/16-svg-irc


          Doug_Schepers, krit, [IPcaller], ed, ChrisL, Tav, cyril,
          cabanier, [Microsoft]




     * [4]Topics
         1. [5]transforms
         2. [6]svg attrs to presentation attrs
         3. [7]process for editing svg2
         4. [8]hamburg f2f
         5. [9]svg2 requirements
         6. [10]The MouseWheel event
         7. [11]Modified sections for hit testing, event processing
         8. [12]XML Events
         9. [13]New Linking features
     * [14]Summary of Action Items

   <trackbot> Date: 16 February 2012

   <scribe> scribenick: ChrisL


   <shepazu> [15]http://dev.w3.org/csswg/css3-transforms/

     [15] http://dev.w3.org/csswg/css3-transforms/

   krit: main differnce is syntax between svg and css, proposed an
   extended parser algorithm to deal with both
   ... erik was in support

   shepazu: could yousummarise how it could be integrated and what the
   changes are?

   krit: we allow more spaces instead of commas, for example

   .com <-- that's the proposal

     [16] http://www.w3.org/mid/8996DBAE-9FE2-44D6-9B9B-B0E4CC05A7E6@adobe.com

   krit: also rotate with three arguments with a coordinate. css does
   not support this they worry about conflict with existing html/css
   ... boris suggested that only svg would allow the three arguments

   shepazu: so for css transforms you specify a transform origin

   krit: no, it affects the whole stack

   ChrisL: so you need to translate, rotate, and translate back

   shepazu: talking to css conent creators, they were frustrated at the
   lack of transform origin for a particular transform

   krit: yes but in a list how would you do that
   ... would need a transform-origin list and it would need to be the
   same length

   shepazu: so the transform origin is different for svg elements

   krit: some images in current spec show how these work together

   Tav: could use more examples

   krit: does not make sense to publish 3d without 2d so we should
   continue with the merged spec
   ... no blocking issues from svg

   shepazu: so current plan of action is to move forward on joint 2d
   and 3d transforms

   krit: 3 point rotate might not be needed for html. parser issues are
   now dealt with
   ... so want a resolution on the parser syntax

   ChrisL: it looked good to me

   (general agreement)

   resolution: adopt dirks' svg parser proposal for transform syntax in

   shepazu: after that is incorporated, we should publish to /TR
   ... change of shortname means its a fpwd despite being published
   ... so sotd needs to explain relationship to previous spec
   ... svg 2d transforms, css 2d transforms (both already published).
   ... so push current spec to fpwd than modify in place the other two
   with obseletion notices
   ... so i propse we resolve to do that

   <ed> s/current spec/current spec -

     [17] http://dev.w3.org/csswg/css3-transforms/

   krit: at CSS WG meeting yesterday it was clear that everyone wanted
   to push forward the merged specification and go to last call on it

   shepazu: yes, so it needs to go to fpwd

   resolution: svgwg agees to publish the css3 merged transform spec as

   shepazu: we should devote a significant amount of css and svg
   telcons to getting this out the door
   ... if there are issues, lets solve them

   tav: status of 3d impl for svg?

   (not yet)

   ChrisL: can still write tests

   shepazu: spec is well enough written that its clear how to write

   tav: there are some unclear places

   shepazu: test help there too

   <scribe> ACTION: erik to inform css wg of our resolution and asking
   them to resolve likewise [recorded in

   <trackbot> Created ACTION-3238 - Inform css wg of our resolution and
   asking them to resolve likewise [on Erik Dahlström - due

   ChrisL: spec has boilerplate autogenerated so it only mentins css wg
   not svg wg. an oversight I am sure

   shepazu: comments should go to fxtf

   cyril: don't want to read all of www-style to get the comments

   shepazu: its easy to follow a link to join a mailing list, for
   people interested.

   krit: css3-transforms keyword in subject

   cyril: but people are not consistent

   shepazu: mailto link can contain the subject "[css3-transforms]"

svg attrs to presentation attrs

   ed: no-one from microsoft here
   ... pinged them but no response

   cyril: we can't wait for them for ever

   krit: spec needs an editor

   ed: we could merge into svg2
   ... if we dont have a main editor for a smaller draft then its
   faster to merge

   krit: can't edit at the moment due to other commitments

   ChrisL: earlier draft had a lot of motivational material and
   rationale which is not needed now we have decided, though it was
   useful early on

   krit: it should be a module because the css transforms spec can
   reference it later

   cyril: ask Brian as he is working on the web animation module
   ... he has worked on common model and syntax and we resolved that in

   rik: myself, brian, shane from google and alex danilo are working on
   this. couple of meetings an hour
   ... wiki pages on w3.org for it

   ed: so we still want a separate module or to put it in svg2?
   ... or make it a separate chapter?

   ChrisL: no opinion either way, just want it to get done

   ed: in fx or in svg?

   ChrisL: unclear

   ed: i have an action to make a draft, was expecting to hand it over
   to patrick. makes sense to have someone from Microsoft as the main
   ... think the spec fits better in svg as its all about svg

   ChrisL: surely if a 1.1 attr is an svg2 property then the
   description needs to be updated

   krit: we have some peoperty names, need to be sure the property type
   is the same as the datatype

   ed: so for general principle, we still need to resolve some issues
   like dom animval. wait for feedback from our developers on that


     [19] http://www.w3.org/Graphics/fx/wiki/SVG_attribute_to_presentation_attribute

   ed: order of stylesheets etc

   krit: graphic is wrong, will republish this evening

   ed: prefer to wait until we get feedback

   krit: also getting feedback on impact on mozilla, webkit, microsoft
   on impact on current implementations. erik your comment is missing

   ed: wil have feedback from dev team by next week
   ... anything else on this topic?

   cabanier: the color-interpolation-filters could wait until next
   week, after getting more feedback

process for editing svg2


     [20] http://lists.w3.org/Archives/Public/public-svg-wg/2012JanMar/0070.html

   <ed> [21]http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Editing_Process

     [21] http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Editing_Process

   ed: so if you own a feature you edit it, main editors czan edit the
   whole document

   cyril: seems to say if you dont own a feature you can't edit it
   ... also good to annotate who owns which bit so its clear who to

   ChrisL: proposal is inconsistent on substantive changes made
   'unilaterally' by the editors

   tav: so i made changes in radial gradients related to border focus
   point. under this i would have to find the feature owner and ask
   them to do it

   cyril: well, you would be the owner in that case

   krit: can there be more than one owner of a section

   ed: over time I'm sure there will be overlap

   Tav: strict rules good at the end but slow us down at the start

   krit; so the main editors are guardians on whether a change should
   be rolled out

   ed: good to have someone to drive a feature and good to have some
   main editors, tav has played that role.

   krit: editorship can change, too

   ed; prefer to have heycam to explain the rationale a bit more

   (krit leaves)

   <cyril> s/amin/main/

   cyril: lets revisit when cameron is here

hamburg f2f

   <ed> [22]http://www.w3.org/Graphics/SVG/WG/wiki/F2F/Hamburg_2012

     [22] http://www.w3.org/Graphics/SVG/WG/wiki/F2F/Hamburg_2012

   ed: pleae register

   roposals <-- you know what to do :)

     [23] http://www.w3.org/Graphics/SVG/WG/wiki/F2F/Hamburg_2012/agenda_proposals

   ChrisL: not got flights (or travel authorisation) worked out yet. i
   see hotels are filling up though

   <vhardy_> Kind reminder to book your hotel. As mentioned on email,
   we only have holds for a few more days.

   ed: arriving same day as brian, staying at clipper

   cyril: as will i

   <vhardy_> vhardy: I will stay at the East

   ChrisL: unclear on date and length of joint meeting

   ed: its the 9th, morning at least

   cyril: not enough for transforms, animations ....

   <vhardy_> vhardy: could make it the whole day?

   ed: need writeupsd and prep for those topics

   +1 for whole day

   ed: plan to have updated filters spec
   ... will confirm with css chairs about ful day

   <scribe> ACTION: erik to check with css chairs that a full day joint
   meeting is okay [recorded in

   <trackbot> Created ACTION-3239 - Check with css chairs that a full
   day joint meeting is okay [on Erik Dahlström - due 2012-02-23].

   ed: are there enough here to fo more svg2 requirements?

svg2 requirements

   cyril: some of them are easy.
   ... lets do the xml events


     [25] http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Requirements_Input#Progress_Events_loadstart.2C_progress.2C_loadend

   ChrisL: is this progress events that j-c duford was editing? does it
   conflict with html5 progress events/

   cyril: use case is download progress bar

   ChrisL: agree on requirement but prefer the html5 solution

   cyril; ok for me


     [26] http://tiffanybbrown.com/2010/07/05/the-html5-video-progress-event/


     [27] http://www.w3.org/TR/html5/the-iframe-element.html#the-video-element

   cyril: looks to be much the same as the svg one. no brainer, adopt

   resolved: accept progress events requirement and adopt html5 method
   for this


     [28] http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Requirements_Input#The_SVGTimer_event

   topic; svg timer event

   ed: think we should drop it. no sense to have an svg specific
   interface for timers

   cyril; is there an html5 equivalent or are people using settimeout

   ed; or request animation frame

   cyril: can do declarative animation and look at the repeat event
   ... fine with dropping it

   resolution svg2 will not have the svg timer event

The MouseWheel event


   ed: reference dom3 evvents for thsi

   ChrisL: is i in dom3 - about to be dropped?

   ed: no
   ... but we need to have the event

   resolved: svg2 will support the mousewheel event as defined in the
   DOM specifications

Modified sections for hit testing, event processing


     [29] http://www.w3.org/TR/SVGMobile12/single-page.html#svgudom-events__MouseWheelEvent

   (we look for links)


     [30] http://www.w3.org/TR/SVGMobile12/single-page.html#interact-PointerEventsProperty

   some stuff on hit testing of text

   ed: and bounding box value
   ... should port over

   reolved: svg2 will adopt improved svgt1.2 text on hit testing and
   event processing

   resolved: svg2 will adopt improved svgt1.2 text on hit testing and
   event processing

XML Events

   ChrisL; drop the entire feature

   resolution: svg2 will adopt improved svgt1.2 text on hit testing and
   event processing

   ed: no need for xml events. could make a seperate module but prefer
   the existing on-* attrs

   cyril: it was a good thing, too bad it was not adopted
   ... what other declarative wat to create or monitor events?
   ... animations plus script

   cyril; and can register a dom listener in script

   scribe: ok to not support it

   resolution: svg2 will not support xml events

New Linking features

   ed: reference restrictions good, needs tweak for svg2 features
   ... different restrictions in 1.1 and 1.2T

   ChrisL: Processing External References is good clarifying text

   uses more up to date language than the sloppy language in 1.1

   ed: yes
   ... again not exactly but a section like that

   resolution: svg2 will adopt the improved wording on Processing
   External References and Reference Restrictions

   ed: needs more work for external references


Summary of Action Items

   [NEW] ACTION: erik to check with css chairs that a full day joint
   meeting is okay [recorded in
   [NEW] ACTION: erik to inform css wg of our resolution and asking
   them to resolve likewise [recorded in

   [End of minutes]

 Chris Lilley   Technical Director, Interaction Domain                 
 W3C Graphics Activity Lead, Fonts Activity Lead
 Co-Chair, W3C Hypertext CG
 Member, CSS, WebFonts, SVG Working Groups
Received on Thursday, 16 February 2012 21:33:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 16 February 2012 21:33:51 GMT