<trackbot> Date: 26 June 2008

<scribe> Scribe: Cameron

<scribe> ScribeNick: heycam

<ed> http://lists.w3.org/Archives/Public/www-svg/2008Jun/0051.html

SVG animate values format

ED: This is about trailing semicolons in the values="" attribute
... and whether you get an additional value if it ends with a ';'

CM: I agree with Julien's assessment that his example should be invalid

DS: Could there be a recovery behaviour here?

<anthony> http://www.w3.org/TR/2005/REC-SMIL2-20051213/animation.html#adef-values

CM: The grammar is defined by SMIL

<ed> http://www.w3.org/TR/SVGMobile12/animate.html#ValuesAttribute

CM: is it by reference?

AG: yes

ED: we could override the definition in SVG

CM: I wonder if there's content out there that uses this semicolon at the end of the list.

<ed> http://lists.w3.org/Archives/Public/www-svg/2008Jun/0020.html

ED: in that mail, andrew points out that the ikivo animator produces content with the semicolon
... and both ikivo and bitflash accept it

AG: what does opera do?

ED: not sure

http://mcc.id.au/temp/semicolon.svg

CM: Batik does accept the semicolon

(discussion about empty string values, like for animating URLs)

CM: you can validly have <a xlink:href=""/>

DS: if we had an image, and we're animating the xlink:href and the last value was the empty string, that would mean that the last value is that the image goes blank

CM: yeah

DS: i'm curious where else SMIL is being used

CM: maybe we can just force the use of '#' for same document references in values=""
... but we should test if empty strings work currently in UAs

<scribe> ACTION: Cameron to test empty string values in animations [recorded in http://www.w3.org/2008/06/26-svg-minutes.html#action01]

<trackbot> Created ACTION-2074 - Test empty string values in animations [on Cameron McCormack - due 2008-07-03].

ISSUE: Behaviour of animating values="a; b; c;" in existing UAs is at odds with the spec

<trackbot> Created ISSUE-2010 - Behaviour of animating values=\"a; b; c;\" in existing UAs is at odds with the spec ; please complete additional details at http://www.w3.org/Graphics/SVG/WG/track/issues/2010/edit .

<ed> http://lists.w3.org/Archives/Public/www-svg/2008Jun/0002.html

SVG Tiny Schema

DS: Chris looked at it a little, and thought maybe it was id vs xml:id
... but Chris says somewhere we introduced id in two different places
... I'm going to be talking next month (maybe before) to the Relax guys, I can ask them about this.

ISSUE: SVG Tiny schema is broken

<trackbot> Created ISSUE-2011 - SVG Tiny schema is broken ; please complete additional details at http://www.w3.org/Graphics/SVG/WG/track/issues/2011/edit .

<shepazu> http://www.w3.org/Graphics/SVG/WG/track/issues/2009

trackbot, close ISSUE-2011

<trackbot> Sorry, heycam, I don't understand 'trackbot, close ISSUE-2011'. Please refer to http://www.w3.org/2005/06/tracker/irc for help

<scribe> ACTION: Doug to figure out why the SVG Tiny schema is broken (ISSUE-2009) [recorded in http://www.w3.org/2008/06/26-svg-minutes.html#action02]

<trackbot> Created ACTION-2075 - Figure out why the SVG Tiny schema is broken (ISSUE-2009) [on Doug Schepers - due 2008-07-03].

ED: reply to felix?

DS: i'll investigate it, see if we can fix it, then reply to felix

Resolving IRIs for resource documents in SVG Tiny 1.2

<ed> http://lists.w3.org/Archives/Public/www-svg/2008May/0033.html

CM: last time we said push it off until we hear from andrew, since bitflash does the double network access already

ED: we don't want to delay spec edits
... maybe we can contact someone else at bitflash

<scribe> ACTION: Cameron to mail some BitFlash guys about ISSUE-2003 [recorded in http://www.w3.org/2008/06/26-svg-minutes.html#action03]

<trackbot> Created ACTION-2076 - Mail some BitFlash guys about ISSUE-2003 [on Cameron McCormack - due 2008-07-03].

percentages in clipPath/pattern/filter/mask content

<ed> http://lists.w3.org/Archives/Public/www-svg/2008May/0048.html

ED: it might be confusing to some people that percentages are relative to the viewport, and not the bbox of the element it's used on
... otoh, if you use values like 0.1, then you get 10% of the object bbox

CM: is it deliberate so that you can use viewport-relative and bbox-relative values in there?

ED: don't know if it's deliberate, but it is confusing

CM: i imagine there might be content out there that uses it

ED: it would be nice to have a test to see if an implementation uses percentages as bbox-relative
... not difficult to change it, but wondering if content relies on it
... first step would be to try implementations
... don't know if it's useful to have the functionality of being able to have a percentage of the viewport
... and at the same time using objectBoundingBox coordinates

<scribe> ACTION: Erik to test implementations for percentage values in clipPath, etc. [recorded in http://www.w3.org/2008/06/26-svg-minutes.html#action04]

<trackbot> Created ACTION-2077 - Test implementations for percentage values in clipPath, etc. [on Erik Dahlström - due 2008-07-03].

ISSUE: Percentages in clipPath/pattern/filter/mask unintuitive

<trackbot> Created ISSUE-2013 - Percentages in clipPath/pattern/filter/mask unintuitive ; please complete additional details at http://www.w3.org/Graphics/SVG/WG/track/issues/2013/edit .

<ed> http://lists.w3.org/Archives/Public/www-svg/2008Jun/0019.html

Tiny 1.2 and Key Identifiers

ED: Is there a way of telling where a keyboard event comes from in the uDOM? I don't think so.
... There's no keyLocation in the tiny KeyboardEvent.

<ed> http://www.w3.org/Graphics/SVG/Group/repository/spec/mobile/1.2/1.2NG/publish/svgudom.html#events__KeyboardEvent

CM: is keyLocation safe in the current DOM3Key spec?

DS: yes

ED: we could add it

DS: it is a subset intended for mobiles, where you only have one keypad

ED: otoh that content might be viewed on different devices

DS: if the content is constrained to the device (if it doesn't assume it can get the different locations) then it doesn't matter
... for uDOM in general, it would be better if we were to add it
... that would change the java bindings

ED: i'd be tempted not to change it in tiny, and in full say that the full DOM3Key interface is required

DS: does that resolve the entire issue?

ED: not really, but part of it
... you'd be saying there's no way to find out where the is coming from, and that you have to rely on DOM 3 Events before being able to decide that
... you could always test if keyLocation is there, and use it if it is
... the modifier keys are missing as well, same thing
... is there any difference in lowercase a and uppercase a? is it the same key identifier?

CM: you'd always get U+0041 i think
... and there's still textEvent yes?

ED: yes
... so you can get the actual typed in string with that

AG: in full we can say to support full DOM3Key

CM: do mobile devices that the udom is targetted to have shift/ctrl keys anyway?

ED: some devices do have shift-like keys

DS: there's IME

ED: mine does have a full(ish) keyboard, with a shift key
... and you can type upper and lowercase letters

<ed> http://www.w3.org/TR/DOM-Level-3-Events/events.html#Events-KeyboardEvent

AG: issue freeze?

CM: you can probably keep a track of shift key state in script if you really needed to

RESOLUTION: We won't add keyLocation and modifier key attributes to the uDOM KeyboardEvent, but we will require the full DOM3Key interface in 1.2 Full

<scribe> ACTION: Doug to reply to Julien to say that we won't add keyLocation etc. to KeyboardEvent in the uDOM [recorded in http://www.w3.org/2008/06/26-svg-minutes.html#action05]

<trackbot> Created ACTION-2078 - Reply to Julien to say that we won't add keyLocation etc. to KeyboardEvent in the uDOM [on Doug Schepers - due 2008-07-03].

Test suite

ED: Has publication happened yet?

DS: no

ED: soon?

DS: yes

ED: we should track test case issues

DS: why not in tracker?

CM: there is a product in there for the test suite

<ed> http://lists.w3.org/Archives/Public/www-svg/2008Jun/0023.html

SVGt 1.2 Tests: Illegal separator in linking-frag-201 and 202

ED: This is using spaces in a URI, and the spec requires using commas without spaces

CM: so escaped spaces (%20) aren't allowed

DS: we could say that the fragment is URL encoded
... if i have a space in my URL now, in a browser, it uses a %20
... we could say if a space is part of a functional fragment then it's escaped

CM: i don't know the intricacies of urls/uris/iris/leiris enough to know
... maybe we can fix the test case to use commas, and raise an issue for the general problem of spaces in there

ED: i agree with kalle's comments on the test description

DS: yes

ED: so that should be changed

<scribe> ACTION: Anthony to fix linking-frag-20[12] and reply to Kalle [recorded in http://www.w3.org/2008/06/26-svg-minutes.html#action06]

<trackbot> Created ACTION-2079 - Fix linking-frag-20[12] and reply to Kalle [on Anthony Grasso - due 2008-07-03].

<ed> http://www.w3.org/Graphics/SVG/Group/repository/spec/mobile/1.2/1.2NG/publish/linking.html#IntroFragmentsViews

ISSUE: The restriction on IRI fragments in view specs to disallow spaces may be overly restrictive

<trackbot> Created ISSUE-2014 - The restriction on IRI fragments in view specs to disallow spaces may be overly restrictive ; please complete additional details at http://www.w3.org/Graphics/SVG/WG/track/issues/2014/edit .

<ed> http://lists.w3.org/Archives/Public/public-svg-wg/2008AprJun/0147.html

XHTML Access module review

<shepazu> http://lists.w3.org/Archives/Public/public-svg-wg/2008AprJun/0161.html

DS: shouldn't take more than half an hour to read through

ED: what is it for?

DS: the idea is that there is an <access> element with a key="" attribute that is a character

CM: this is for keyboard shortcuts?

DS: yes
... right now we have the accesskey="" we got from SMIL, but that is known to be a problem
... you map the values of the <access> element to things in the DOM with targetid=""
... or you can use targetrole="", which will make the access key go to the element with that role
... you can have lists in these attributes, which will cause cycling through when you press the access key multiple times
... if you say activate="yes", then if the user presses the key, it will shift the focus to that element and activate it (e.g. if it's a link it will follow it)
... it'd be a good part of the role stuff we'll have in svg
... what if you wanted to describe a sequence of things that weren't in document order? e.g. when you hit F it goes to the first node, then hitting F goes to the second node, which may or may not be in document order

ED: seems strange to be document order, and not order in the attribute
... has to be one key, not several keys? e.g. if you are used to emacs shortcuts you can't add those in addition to your regular keys?
... or modifier keys

DS: you could have both P and Q mapped to the same element with multiple <access> elements

ED: can't do multi-key-sequences either, e.g. M followed be E

DS: an interesting example, not sure if it's a powerful use case

ED: encourage you all to send comments

DS: due date is already past, but someone asked for an extension so if we do them by next week it would be great

This is scribe.perl Revision: 1.133  of Date: 2008/01/18 18:48:51  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/look at requiring/require/
Found Scribe: Cameron
Found ScribeNick: heycam

WARNING: No "Present: ... " found!
Possibly Present: AG CM DS Doug_Schepers IPcaller ISSUE P4 P5 ScribeNick anthony ed ed_ heycam joined shepazu svg trackbot
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy

Regrets: Andrew_Emmons Chris_Lilley
Agenda: http://lists.w3.org/Archives/Public/public-svg-wg/2008AprJun/0162.html
Found Date: 26 Jun 2008
Guessing minutes URL: http://www.w3.org/2008/06/26-svg-minutes.html
People with action items: anthony cameron doug erik

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