SVG Working Group Teleconference

05 Oct 2010

See also: IRC log


[Microsoft], ed, +39.537.7.aaaa, Shepazu, [IPcaller]


<trackbot> Date: 05 October 2010



<trackbot> ISSUE-2353 -- Last Call Comment: Clarify coordinates for feImage. -- raised

<trackbot> http://www.w3.org/Graphics/SVG/WG/track/issues/2353



SVG Integration

<scribe> scribeNick: ed

DS: AG updated the tables, some minor bugs but it's good enough as a starting point
... added elements, attributes and properties into SVGIntegration
... and filed a bug with HTML5 to not whitelist some attributes and elements but rather to refer to SVG Integration

<shepazu> http://www.w3.org/Bugs/Public/show_bug.cgi?id=10965

DS: we've discussed this concern before
... HTML5 will be contemporary with SVG2
... need to collect all attributes and elements from all specs
... i've outlined an algorithm for dealing with mixed-case
... it's well established
... in unicode

PD: that's lowercasing everything, right?
... could you forward-compatible this, such that elements automatically work?
... can we skip the tables?

DS: casematching is an issue, most authors don't care, but xml does
... but svg in html isn't xml anymore
... could consider it as two serializations
... or we could make schema that allows both
... not sure it's worth it
... though if we think people will screw up casing
... have seen people getting 'viewBox' wrong, it has a capital B, i've seen 'viewbox' many times

PD: don't buy your xslt argument, if you want to use that you make sure you have valid xml

DS: we'll face similar issues with CSS properties, where properties are shared
... e.g with lowercasing case-sensitive values

PD: there are technolgies that were done in svg first, but which are now moving into the html5 space
... things that aren't necessarily bound to vector graphics

DS: we should try to make features maximally useful to the web languages
... svg didn't define the xml format, wouldn't mind if there was a similar effort to make xml-like format that allows error-correction as html5 does, such that we get case-insensitivty, leaving off attribute quotemarks
... maybe there should be a more forgiving parser for xml
... in the svg2 timeframe, we need to resolve the case sensititvy

PD: we need to take a stand on strict xml vs looser syntax
... it's the most important topic

DS: PD you're writing up a whitepaper for tpac?

PD: more of a gathering of thoughts

<scribe> ...in progress, trying to get it ready before tpac

UNKNOWN_SPEAKER: to get feedback on it

DS: as a starting position that would be useful

PD: we need to find our position, xml or not

DS: the goal is to adopt svg, not particulary xml
... the case sensitityv is an obstacle

PD: people are likely going to learn svg from the html perspective
... they're doing css transforms, not svg transforms

DS: we should put a high priority on filters, and transforms
... anything we are going to be sharing

<jwatt> Zakim: ++P0 is me

<jwatt> Zakim: ??P0 is me

DS: how do we make it as easy as possible for people to use svg? with the ideoms they're already familiar with
... PD said that he thinks this is urgent

new zeeland f2f

PD: when's that?

DS: late february

ED: JW do you have any more details?

JW: need to narrow down the dates

DS: could you send out an email?

JW: ok

DS: PD will you be able to go?

PD: will check


PD: how do we get the microsoft tests accepted by the wg?

ED: get them reviewed by people on the group, know it's hard and people are pressed for time and so on

PD: we are working on categorizing the tests better, to say that the users can code to particular modules
... we need to get the message across that svg is ready for primetime

DS: when i did presentations filters is the thing that excites people the most
... maybe it will come to be seen as a css feature though
... it's still percieved as an svg feature
... we might start talking about reorganization, functionality sets
... we're doing a report on the testsuite, on tests that we think are broken
... some tests are using features like declarative animation
... our proposal would be to make the tests more finegrained
... the test naming
... an alternative route... a dependency chain for each spec
... for each test what part of the spec is that test testing
... in terms of organizing the tests we should add metadata to say what dependencies the test has
... so if you have filters in a markers test, link to those sections

PD: sounds like a lot of work
... there are too many dependencies in the tests

DS: we could categorize the tests, renaming is one thing

ED: renaming tests in cvs, not a great idea, loses history
... would love to move to more modern version control system though

PD: beauty of the web website has some great svg examples

<pdengler> http://www.beautyoftheweb.com/

<pdengler> http://www.thekillersmusic.com/HTML5#tour-browse

<pdengler> http://www.skybeautiful.com/


<shepazu> http://lists.w3.org/Archives/Public/www-svg/2010Oct/0031.html

ED: sounds a bit like feComponentTransfer

TB: this is indexed colors

PD: let's put it in the issue tracker

DS: i'd like to cut out all but one color channel
... e.g take away all color except blue
... is that doable?

PD: is that different from masking?

DS: don't know

trackbot, make minutes

<trackbot> Sorry, ed, I don't understand 'trackbot, make minutes'. Please refer to http://www.w3.org/2005/06/tracker/irc for help

trackbot, end telcon

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/10/05 21:38:16 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
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: ed
Inferring Scribes: ed
Default Present: [Microsoft], ed, +39.537.7.aaaa, Shepazu, [IPcaller]
Present: [Microsoft] ed +39.537.7.aaaa Shepazu [IPcaller]
Regrets: CL
Found Date: 05 Oct 2010
Guessing minutes URL: http://www.w3.org/2010/10/05-svg-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]