See also: IRC log
<trackbot> Date: 13 September 2011
<scribe> scribe: davelab6
<scribe> scribenick: davelab6
how to scribe pls? :)
scribenick
scribe davelab6
<ChrisL> scribenick: davelab6
Vlad: We're going to tabk about the Conformance Test Suite
<ChrisL> chris displays http://w3c-test.org/framework/review/woff-ua/
<ChrisL> blocks extraneous* failsin all browsers
<ChrisL> blocksoverlap tests fail inmostbrowsers
<ChrisL> metadata tests give free pass to implementationsthat dont displaymetadata
<ChrisL> (seeemail fron jfkthame)
ChrisL: If the extended metadata block is ignored, then the tests pass, but if displaying it to users is implemented but not correctly, it will fail, which incentivises UA developers to not implement showing metadata to users. So we'd like to change this so that implementing this optional part is incentivised.
Vlad: Also change the colors so that red means 'no test' (ie, no implementation), yellow means 'test fails' and green is 'test passes'
<ChrisL> resolved: change metadata tests so only implementationsthatdisplay metadata are tested
<ChrisL> ACTION: chris to fix metadata* results for free passes [recorded in http://www.w3.org/2011/09/13-webfonts-minutes.html#action01]
<trackbot> Created ACTION-103 - Fix metadata* results for free passes [on Chris Lilley - due 2011-09-20].
<ChrisL> ACTION:tab to update test suite for metadata tests [recorded in http://www.w3.org/2011/09/13-webfonts-minutes.html#action02]
<ChrisL> we should pressure implementors to fix code for the overlapping blocks
cslye: UA should ignore the data, and if they read it likely to fail other tests
ChrisL: should we keep spec as it is and ask implementors to fix, or change spec?
cslye: in 2 years we dont want a security advisory out about this
vlad: not sure if this could be a real security risk?
tab: if directory is set incorrectly, is that our concern? sfnt table size could be 4 bytes larger than it is...
vlad: you'd try to get extra data inside a table so its loaded, not insert data between tables
<ChrisL> header-totabSfntSize* are failing in some browsers but we have enough passes for CR. Webkit has a pass (Chrome) and a fail (Safari) due to the googlesanitiser,we suspect
tab: we should put a metadata display into the validator as an option
<ChrisL> (tab checks whetherW3C Unicorn validation framework canoutput purely informative material - yes)
<ChrisL> ACTION: chris to check on unicorn integration of the woff validator [recorded in http://www.w3.org/2011/09/13-webfonts-minutes.html#action03]
<trackbot> Created ACTION-104 - Check on unicorn integration of the woff validator [on Chris Lilley - due 2011-09-20].
chris: this will allow people to
submit a WOFF and get it validated, and as part of the
validation report the user can have it display the optional
metadata
... break for lunch, and try and find tab :)
<ChrisL> ACTION: tab to check whether Chromeexposesan API to extendsions that shows the specific font in use [recorded in http://www.w3.org/2011/09/13-webfonts-minutes.html#action04]
<trackbot> Created ACTION-105 - Check whether Chromeexposesan API to extendsions that shows the specific font in use [on Tab Atkins Jr. - due 2011-09-20].
<ChrisL> vlad:once unicorn has the validator we will be in good tshape there
<ChrisL> tal:and add the metadata too; need to check if info not related to validity is ok
<ChrisL> scribenick: ChrisL
ChrisL: we have jfkthame script and fontforge, what else
Vlad: tests need to be entered manualy
ChrisL: yes
... is fontsquirrelusing their own implementation? (we
don'tknow)
tal: wecan test mywoff toolsas well
cslye: does the fontlab beta export woff now?
tal: this is not a listed feature in their betas
(digression on epar)
http://www.eeulaa.org/index.php?option=com_content&view=article&id=22&Itemid=34
ChrisL: online font converter http://onlinefontconverter.com/
<davelab6> scribenick davelab6
<davelab6> ChrisL: onlinefontconverter.com has a WOFF converter, we will test it now
<davelab6> davelab6: stefan.warting@gmail.com is the email in the whois
ok so it seems to convert sfnt to woff (incliding the invalid ones)
cslye: interesting to ask typekit and extensis webink what they are usiing
tal;and other services
davelab6: phil taylor did sometools in perl
<scribe> ACTION:chris to run authoring tools against the uatests and make report [recorded in http://www.w3.org/2011/09/13-webfonts-minutes.html#action05]
<scribe> ACTION: chris to run authoring tools against the uatests and make report [recorded in http://www.w3.org/2011/09/13-webfonts-minutes.html#action06]
<trackbot> Created ACTION-106 - Run authoring tools against the uatests and make report [on Chris Lilley - due 2011-09-20].
Vlad: can we talk to the author of the online converter and get them to fix bugs?
ChrisL: i can ask
<davelab6> ChrisL: so we've done a lot today, we have plans to move forward and actions assigned to peopl
<davelab6> Vlad: agreed, i wonder about a TPAC meeting, would be good to move forward. tab, will you be there?
<davelab6> tab: yes
<davelab6> vlad: the same origin restriction, it will being web fonts and web apps (?) groups together
<davelab6> tab: where will it be?
<davelab6> ChrisL: bay area
<davelab6> cSlye: ill be there
<davelab6> Vlad: early bird discount
http://www.w3.org/2011/11/TPAC/
14 October - Registration Closes and Registration fee goes up
(adjourned)
This is scribe.perl Revision: 1.136 of Date: 2011/05/12 12:01:43 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/tal/tab/g Found Scribe: davelab6 Inferring ScribeNick: davelab6 Found ScribeNick: davelab6 Found ScribeNick: davelab6 Found ScribeNick: ChrisL ScribeNicks: davelab6, ChrisL Present: davelab6 tal sergeym Zakim RRSAgent ChrisL jfkthame_afk Vlad tab WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 13 Sep 2011 Guessing minutes URL: http://www.w3.org/2011/09/13-webfonts-minutes.html People with action items: chris tab WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]