See also: IRC log
<trackbot> Date: 02 February 2009
<scribe> scribe: erik
<scribe> scribeNick: ed_
<ChrisL> issue-2201?
<trackbot> ISSUE-2201 -- Return value of SVGAnimationElement.getStartTime unclear -- CLOSED
<trackbot> http://www.w3.org/Graphics/SVG/WG/track/issues/2201
ED: question is whether the
decided upon behaviour was the most useful to authors
... have implemented the resolution we came to
CL: seems resonable to raise exception if you cant calculate the starttime
<ChrisL> It seems reasonable to me. if the re is no current interval, then throw an exception
CMC: you can determine whether
it's the current starttime or the previous interval
starttime
... don't mind either way
... CL if you're concerned about whether it's not
current...
CL: right, but it's only about whether a begin time could be determined
ED: is it possible for an author to get the previous starttime?
CMC: perhaps by listening to endEvents
CL: is there information that the impl has that can be given back by the exceptioncodes
CMC: currently just trhows invalid_access_err
ED: the counter-argument for listening to endEvents is listening to beginEvents
CMC: for either way yes, the workaround is to listen for events
ED: ok, at least there are
workarounds
... fine, the current errata text will be fine
<scribe> ACTION: heycam to do the suggested changes/additions to the http://dev.w3.org/SVG/profiles/1.1F2/errata/errata.xml#getStartTime-return-value errata item, see http://lists.w3.org/Archives/Public/www-svg/2009Jan/0050.html [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action01]
<trackbot> Created ACTION-2430 - Do the suggested changes/additions to the http://dev.w3.org/SVG/profiles/1.1F2/errata/errata.xml#getStartTime-return-value errata item, see http://lists.w3.org/Archives/Public/www-svg/2009Jan/0050.html [on Cameron McCormack - due 2009-02-09].
<heycam> ISSUE-2071?
<trackbot> ISSUE-2071 -- potential security hole involving pointer-events, filters, foreignObject, cross-origin IFRAMEs, and elementFromPoint -- RAISED
<trackbot> http://www.w3.org/Graphics/SVG/WG/track/issues/2071
CMC: Robert sent some examples of how one might be able to access crossorigin content
CL: it's using filters to access a part of the screen that you own
<ChrisL> Opera has getPixel and setPixel on canvas, might be another way without filters
CMC: right, mozillas blingthing
had something similar IIRC
... I think it should probably be tracked per-origin
CL: if you have an svg that
brings in a raster image from somewhere else, then you don't
want your filters to stop working
... with some strange securityviolation error
CMC: the fact that filters can
use the image is ok, but you might have to restrict how scripts
can access that information
... some ppl have experimented wrapping entire sites in some
color tint filters, and that sort of things should still
work
... the particular thing is pointer-events=painted, maybe the
wording could be tweaked to cover that hole
... but it's more of a general problem
CL: exposing pixel values to script is the issue
CMC: one way would be to track
pixel per pixel the origin
... the other might be to [fill-in-blank]
<heycam> s/[fill-in-the-blank]/have a single "tainted" flag that gets set whenever foreign content is accessible via a particular canvas, etc./
<heycam> ...which would prevent script from accessing any of the pixels there
CMC: ED did you want to ask your security guys about this?
ED: I think for this particular issue that's raised we weren't vulnerable, but I'll inquire about the other possible exploits accessing pixel values
CMC: might be good to discuss at the F2F
<scribe> ACTION: ED to start the discussion on security SVG/canvas/filters etc with the security-group at opera [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action02]
<trackbot> Created ACTION-2431 - Start the discussion on security SVG/canvas/filters etc with the security-group at opera [on Erik Dahlström - due 2009-02-09].
ISSUE-2026?
<trackbot> ISSUE-2026 -- Need Test Ensuring that 'class' is Animatable -- RAISED
<trackbot> http://www.w3.org/Graphics/SVG/WG/track/issues/2026
CMC: the rest are issues to create tests, they jsut don't have actions yet
CL: on svg full 1.2 testsuite, but should probably be 1.1 testsuite
CMC: though we might be able to getPresentationTrait in 1.2 full
CL: seems like an easy test to write
<scribe> ACTION: CL to create an SVG 1.1 test verifying that the class attribute is animatable [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action03]
<trackbot> Created ACTION-2432 - Create an SVG 1.1 test verifying that the class attribute is animatable [on Chris Lilley - due 2009-02-09].
<scribe> ACTION: CL to create an SVG 1.2 full test verifying that the class attribute is animatable, relates to ISSUE-2026 [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action04]
<trackbot> Created ACTION-2433 - Create an SVG 1.2 full test verifying that the class attribute is animatable, relates to ISSUE-2026 [on Chris Lilley - due 2009-02-09].
CMC: this sounds like a tricky one to test
CL: not really
... use cursorkeys and normal keys
... put on capslock and shift, that could be done as a
sequence
... though it might be hard if you have a device that don't
have those keys
CMC: ought to be comprehensive,
maybe a list of keys that we know don't generate any text
... doesn't seem particularly urgent
<scribe> ACTION: heycam to create 1.2T tests for ISSUE-2018, ISSUE-2019, ISSUE-2020 [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action05]
<trackbot> Created ACTION-2434 - Create 1.2T tests for ISSUE-2018, ISSUE-2019, ISSUE-2020 [on Cameron McCormack - due 2009-02-09].
<heycam> http://dev.w3.org/SVG/profiles/1.1F2/errata/errata.xml#xml_space
<ChrisL> action-2165?
<trackbot> ACTION-2165 -- Erik Dahlström to add a 1.1 errata item for having xml:space on tspan elements to align with 1.2T (split out from ACTION-2048) -- due 2008-08-30 -- CLOSED
<trackbot> http://www.w3.org/Graphics/SVG/WG/track/actions/2165
ED: basically just aligning with
1.2T, minor clarification
... cat3 correct?
CL: seems like a cat3 yes
ED: can we move it to proposed?
CMC: yes, looks
... fine
<heycam> <span class='attr-name'>'xml:space'</span>
<heycam> <a href=...><span class='element-name'>'text'</span></a>
<scribe> ACTION: fix markup for http://www.w3.org/Graphics/SVG/WG/track/actions/2165 as indicated in these minutes and move to proposed [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action06]
<trackbot> Sorry, couldn't find user - fix
<scribe> ACTION: ed to fix markup for http://www.w3.org/Graphics/SVG/WG/track/actions/2165 as indicated in these minutes and move to proposed [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action07]
<trackbot> Created ACTION-2435 - Fix markup for http://www.w3.org/Graphics/SVG/WG/track/actions/2165 as indicated in these minutes and move to proposed [on Erik Dahlström - due 2009-02-09].
<ChrisL> action-2397?
<trackbot> ACTION-2397 -- Erik Dahlström to add 'loadstart', 'progress' and 'loadend' as animation events to the 1.2T errata -- due 2009-01-19 -- CLOSED
<trackbot> http://www.w3.org/Graphics/SVG/WG/track/actions/2397
<heycam> http://dev.w3.org/SVG/profiles/1.2T/errata/errata.xml#progressevents
ED: should probably fix the title of it
CMC: yep, looks ok, move to proposed
CMC: need some markup for xlink:href there
AG: looks good to me
<scribe> ACTION: ed to fix the markup and move the two errata items to proposed: http://dev.w3.org/SVG/profiles/1.2T/errata/errata.xml#progressevent_eventtarget and http://dev.w3.org/SVG/profiles/1.2T/errata/errata.xml#progressevents [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action08]
<trackbot> Created ACTION-2436 - Fix the markup and move the two errata items to proposed: http://dev.w3.org/SVG/profiles/1.2T/errata/errata.xml#progressevent_eventtarget and http://dev.w3.org/SVG/profiles/1.2T/errata/errata.xml#progressevents [on Erik Dahlström - due 2009-02-09].
AG: we're waiting for some actions on zoomevents (on jwatt)
AG: there are some items on doug and some on jwatt, but we should try to get those sorted at the f2f
CMC: not sure which ones need to be done before the f2f
<heycam> ISSUE-2113?
<trackbot> ISSUE-2113 -- animate-elem-35 -- RAISED
<trackbot> http://www.w3.org/Graphics/SVG/WG/track/issues/2113
CMC: comment from someone saying
that one of the 1.2T tests has a problem because it relies on
particular precision
... someone should probably take action to analyze this
CL: would it help having an
explicit lineto instead of the closepath?
... also don't understand why it doesn't amp to fixedpoint
coordinates
CMC: might be that the test could
be tweaked
... would need someone to go through the calculations
CL: do we know if any of the mobiles use fixedpoint?
CMC: I think so
... if the calculations come out fine in 16.16 we can probably
close the issue
... think the mail there explains why they chose something else
than 16.16
CL: what's the company?
CMC: ekioh
... maybe we should restructure the test to run inside some
given precision
... we should know from the implementation report which
implementations pass
ED: though all the implementations were run on desktop, might be different on device
<scribe> ACTION: AG to see if the animate-elem-35 can be remade to work under smaller precision fixedpoint [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action09]
<trackbot> Created ACTION-2437 - See if the animate-elem-35 can be remade to work under smaller precision fixedpoint [on Anthony Grasso - due 2009-02-09].
<heycam> ISSUE-2179?
<trackbot> ISSUE-2179 -- text-fonts-202-t has wrong reference image -- RAISED
<trackbot> http://www.w3.org/Graphics/SVG/WG/track/issues/2179
CMC: probably only to regenerate the reference image
<scribe> ACTION: CL to fix the text-fonts-202-t reference image, addressing ISSUE-2179 [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action10]
<trackbot> Created ACTION-2438 - Fix the text-fonts-202-t reference image, addressing ISSUE-2179 [on Chris Lilley - due 2009-02-09].
CMC: does tiny 1.2 require external fonts?
CL: yes, but only svg fonts are
required
... will make an svgfont of the font (have permission to do
so)
<scribe> ACTION: CL to convert the text-fonts-202-t truetype font to an svgfont and use for another test [recorded in http://www.w3.org/2009/02/02-svg-minutes.html#action11]
<trackbot> Created ACTION-2439 - Convert the text-fonts-202-t truetype font to an svgfont and use for another test [on Chris Lilley - due 2009-02-09].
thursday topics: filters, compositing
scribe: david daileys suggestions
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/argument/counter-argument/ FAILED: s/[fill-in-the-blank]/have a single "tainted" flag that gets set whenever foreign content is accessible via a particular canvas, etc./ Found Scribe: erik Found ScribeNick: ed_ Default Present: heycam, ChrisL, Erik, anthony Present: heycam ChrisL Erik anthony Found Date: 02 Feb 2009 Guessing minutes URL: http://www.w3.org/2009/02/02-svg-minutes.html People with action items: ag cl ed fix heycam[End of scribe.perl diagnostic output]