SVG Working Group Teleconference -- 19 Apr 2012

<trackbot> Date: 19 April 2012

I try, but I can not call in. The US number doesn't work for me

heycam: I don't:(

<Cyril> it is working for me via Skype

<cabanier> for me too

<ChrisL> tav, speak for a moment

<heycam> Scribe: krit

Progress update on SVG 2

<heycam> http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Requirements_Commitments

heycam: Want to see the assignments on the wiki
... original plan was publish 1st WD at the end of the month
... will be sections added we assgined for
... + changes

Cyril: it is a chnage to the planning on the page

<Cyril> http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Planning_Page

Cyril: all placeholdershould be published by april
... we have to prepare for traveling and so on

heycam: we won't publish the WD in april

Cyril: what about july? too late?

heycam: it depends
... published draft even very intial would be good
... who did not so far, sign up till next week
... and add topics into spec

shepazu: is playback for audio element? script API?

heycam: let me take a look

<heycam> http://www.w3.org/Graphics/SVG/WG/track/issues/2069

heycam: proposals

shepazu: markup properties or attrubutes?
... I m not an audio guy but in the audiio WD. I won't work on it ut am happy to talk with the people there.
... we should coordinate with audio WG

heycam: move on

shepazu: one more question
... I put a question mark on one of the list items
... In case someone wants to work on it
... just contact on IRC

heycam: even with question marks, looks like you are assigned

shepazu: will we have a confilct

heycam: was off topic … confilcts on wiki page

<ChrisL> I was editing (and saving) a bunch during the call

heycam: next topic

SVG 2 editing process

heycam: I came up with a proposal

ChrisL: I think it is fine in general
... we can discuss if some one to objects

<heycam> http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Editing_Process

shepazu: It changes to resolved

heycam: people asgined to a feature go ahead to change the section they are asigned for. Group will check it afterwards

ChrisL: is the list is complete? I don't find sths, like mask (alpha, luminance)
... I'd like to assign for it

krit: Vhardy and ed o'connnor want to work on it on FX TF

ChrisL: right, but how does it get into SVG 2

<ChrisL> ok as long as someone is doing it. css wg seems to want this urgently

<heycam> http://www.w3.org/Graphics/SVG/WG/track/issues/2401

<Cyril> http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Requirements_Input#Hit-testing_on_image_alpha

<ChrisL> cyril that is not the same

ChrisL: people say we want what svg does
... but it does more than we want

Cyril: can you explain

ChrisL: currently masking use the mask element
... mask is done by calculating luminance and that gets multiplied with the alpha
... we may want just have alpha masking
... don't find resolutions for it

<ChrisL> ACTION: chris propose an update to masking for luminance and alpha and pointing directly at images [recorded in http://www.w3.org/2012/04/19-svg-minutes.html#action01]

<trackbot> Created ACTION-3256 - Propose an update to masking for luminance and alpha and pointing directly at images [on Chris Lilley - due 2012-04-26].

<scribe> ACTION: ChrisL will look at alpha masking and luminance masking and talk to vhardy [recorded in http://www.w3.org/2012/04/19-svg-minutes.html#action02]

<trackbot> Created ACTION-3257 - Will look at alpha masking and luminance masking and talk to vhardy [on Chris Lilley - due 2012-04-26].

<Cyril> http://www.w3.org/Graphics/SVG/WG/wiki/SVG2_Requirements_Input#Late_Proposed_Requirements

Cyril: we should take a look at the late propoesed requirementss

<scribe> scribeNick: heycam

<ChrisL> scribenick: chrisl

Tav: can wee resolve on the requirement for masking

(yes)

resolution: we will allow masks to use either alpha or luminance or both

ChrisL: and pointing direct to images?

tav: how would you specify if you want alpha or what

ChrisL: the property would do that

resolution: : svg will relax restriction on masks pointing to mask element only

Cyril: did we resolve on the editing proposal

heycam: yas and we already resolved on it earlier apparently

ChrisL: prefer to soften the three week paer

tav: yes

heycam: ok

test content guidelines

heycam: brian was aksing for guidelines so they can succeed or fail in the same way like all green rectangles

tav: like red and green but if they all look the same it is confusing
... doesnt say what is being tested
... dont want a rect that covers the whole thing

heycam: if the tests are automated you won't need to look at them

Tav: so there is no debugging help

Cyril: point is to automate error detection

heycam: also easier to make reference images. but with text or paths there can be pixel differences. also easier to determine pass or fail

Tav: but for transforms, css or attr version should be te same so no pixel diffferences

heycam: ok
... some tests might be simple enough you want just a plain rect.

Tav: ok for red gren fail/pass unless it is testing colors

shepazu: we need to test for red colours for example

Tav: we did that on the last test suite

shepazu: good to have tests that can be automatically detected

Cyril: yes bt we do not need green flood fill for that

Tav: subtract the ref images

shepazu: ok

heycam: for animation, brian suggests tests where the final state is at one second so there is a snapshot to compare to the ref
... script can sett the time to that point

ChrisL: for path animations you need multiple snapshots surely

heycam: we agreed to add snapshot time so we could use that to annotate the test

Cyril: gpac has a way to generate an xml which logs the snapshot time and you can have key events and when it replays, the snapshots are generated at the marked times
... want it sampled every time there is a change. the author knows the critical times

heycam: yes it is really a feature of the test itself
... avoid tests with many simultaneous animations

tav: examples of animation reftests?

heycam: not yet

Cyril: can send an example

<Tav> http://tavmjong.free.fr/SVG/CSS/Transforms/transforms-basic-001.html

<scribe> ACTION: cyril to send examples of gpac animation testing [recorded in http://www.w3.org/2012/04/19-svg-minutes.html#action03]

<trackbot> Created ACTION-3258 - Send examples of gpac animation testing [on Cyril Concolato - due 2012-04-26].

heycam: like that excample, tav

tav: no browser passes yet

heycam: ok so aiming for a single green or red rect is not good, but if it is a simple pass/fail result then go for the rect

ChrisL: sometimes you can loose track if all the tests look the same

(general agreement)

heycam; does someone have an action to set up the test structure

tav: i set it up like css. each contributor has a submit directory
... once apprved they get moved
... the structure to run through the reftests is not done

heycam; wonder how easy t is to handle animation tests

tav: we talked with peter Linss who said it was possible. we need an example

<heycam> ScribeNick: heycam

heycam: would be good to have these test guidelines on a wiki page

Tav: we also have the test template that I worked on

… we resolved on having a link to a copyright license, waiting on chris for that

<scribe> ACTION: Cameron to create a test guideline page, mentioning rectangles and linking to template file [recorded in http://www.w3.org/2012/04/19-svg-minutes.html#action04]

<trackbot> Created ACTION-3259 - Create a test guideline page, mentioning rectangles and linking to template file [on Cameron McCormack - due 2012-04-26].

F2Fs

Tav: I will probably be going to SVG Open

heycam: I am not sure whether we want to meet at both SVG Open and TPAC

<ChrisL> scribenick: ChrisL

ChrisL: (describes how to set up the css harness)

tav: can easily make a script that extracts that

<heycam> ACTION: Tav to work with Chris to get test suite manifest scripts working [recorded in http://www.w3.org/2012/04/19-svg-minutes.html#action05]

<trackbot> Created ACTION-3260 - Work with Chris to get test suite manifest scripts working [on Tavmjong Bah - due 2012-04-26].

<scribe> ACTION: tav work with chris to create the test manifest [recorded in http://www.w3.org/2012/04/19-svg-minutes.html#action06]

<trackbot> Created ACTION-3261 - Work with chris to create the test manifest [on Tavmjong Bah - due 2012-04-26].

<Tav> http://www.w3.org/Graphics/SVG/WG/wiki/SVG2/Testing_Requirements

tav: chris we still lack the copyright link

ChrisL: oh yes had forgotten about that

f2f

heycam: lets not have 5 meetings

Cyril: is the seattle one necessary

shepazu; 5 is a lot

Cyril: how about a seattle-paris videoconferenced meeting?

Tav: its 9 hours difference. doable

Cyril: might have some videoconf. polycom thing.

cabanier: yes we have that too

Cyril: will not be able to make seattle

shepazu: maybe microsoft has facilities that would host this

jen: yeah we could look into that

25-27 july

cabanier: adone is hosting but no issue if microsoft hosts

Cyril; can tell you in a couple of weeks in paristech can host

ChrisL; a split paris-seattle would be good for me

ChrisL; would prefer to meet at svgopen rather thn tpac if we have to choose

shepazu; who is going to tpac anyway

(4 or 5)

shepazu: so maybe a short mostly-liaison meeting at tpac

heycam: ok so tell tpac that we only need one day?

resolved: one day liaison meeting at tpac

resolved: split seattle/paris ftf in july if feasible

resolved: meet at svgopen/graphicalweb in zurich, main meet rather than tpac

heycam: three meetings 2 months apart

ChrisL: but fits in well with svg2 editing push

review of path related cancas additions

<heycam> http://lists.w3.org/Archives/Public/public-svg-wg/2012JanMar/0192.html

ChrisL: dmitry had a nice proposal with a path-like syntax for animations

heycam: relevat for the improved dom work
... as for feedback to html, my comments in the mail are that its is generally ok
... but how the object relates to our pathseglist which is yucky

cabanier: idea is that people share knowledge between using canvas and svg

heycam; pathseglist can't assign a new object, need to either replace all items or set upa whole new path data stream

scribe: so for dom improvement, we should improve path and that informs how we relate to the html canvas path api. need to lok at our improvements before deciding that
... use of svgmatrix, functions for path to stroke or svg text on path are all fine as is

shepazu: any other path based operations

heycam; can't get path data out. can only paint it. its opaque

shepazu: want an api that gives you vector effects like stuff

ChrisL: yes we said we would add that on vector effects

ChrisL; are the opaque path objects a problem here or not/ they can be manipulated, just not inspected

heycam; might be a simplification or might be a way to avoid exposing it to script

cabanier: gives more freedo on internal storage
... often you just want the result

shepazu: can you do it with text and thus capture the text

heycam: yes
... we wanted to capture text as outlines

shepazu: so are these in the svg2 requirements

heycam: dont think

resolution: we will have a vector effects api to get reesulting paths (ned to discuss normalisation)

ChrisL: easier to implement if the generated paths are not standardised, onkly the shape they make

heycam: danger of a small, deep rathole
... so for review comments

(general agreement with hecam's suggestions)

<scribe> chair: heycam

<heycam> ACTION: Cameron to mail HTML WG with path API comments [recorded in http://www.w3.org/2012/04/19-svg-minutes.html#action07]

<trackbot> Created ACTION-3262 - Mail HTML WG with path API comments [on Cameron McCormack - due 2012-04-26].

- DRAFT -

SVG Working Group Teleconference

19 Apr 2012

Attendees

Contents

Progress update on SVG 2

SVG 2 editing process

test content guidelines

F2Fs

f2f

review of path related cancas additions

Summary of Action Items

Scribe.perl diagnostic output