See also: IRC log
<trackbot> Date: 04 August 2010
<scribe> Scribe: ChrisL
Vlad: Thanks jdaggett for reviewing the spec
http://lists.w3.org/Archives/Public/www-font/2010JulSep/0027.html
jdaggett: most of these are
editorial
... two major things
... section 8 conformance, checksum must be correct,
implication is that checksum is computed and checked,
... must either correct the values or issue a warning
... need to be consistent, reject or warn?
... many checks need to happen but that is not part of woff,
hapens once a font is converted
sergeym: validating a checksum is an additional step
jdaggett: not clear if we can reuse the header checksum or required to recompute it
sergeym: UA does not change table
structure
... so the checksum is still valid
jdaggett: wording says the tool should validate it
Vlad: sergeym seems to be agreeing that checksum validation should not be required of UA
resolved: UAS need not validate checksums
<scribe> ACTION: jjonathan to clarify that checksums are not checked by a user agent [recorded in http://www.w3.org/2010/08/04-webfonts-minutes.html#action01]
<trackbot> Sorry, couldn't find user - jjonathan
<scribe> ACTION: jjonothan to clarify that checksums are not checked by a user agent [recorded in http://www.w3.org/2010/08/04-webfonts-minutes.html#action02]
<trackbot> Sorry, couldn't find user - jjonothan
<scribe> ACTION: jonothan to clarify that checksums are not checked by a user agent [recorded in http://www.w3.org/2010/08/04-webfonts-minutes.html#action03]
<trackbot> Sorry, couldn't find user - jonothan
<scribe> ACTION: jonathan to clarify that checksums are not checked by a user agent [recorded in http://www.w3.org/2010/08/04-webfonts-minutes.html#action04]
<trackbot> Created ACTION-12 - Clarify that checksums are not checked by a user agent [on Jonathan Kew - due 2010-08-11].
jdaggett: better to move the conformance requirenents into section 8, list for tools and user agents
Vlad: yes appendices should be
informative so notrmative text should be moved out of
there
... of if it looks normative bit isnt it should be reworded
jdaggett: structural issue,
repeating similar ideas in 2 parts of a document they get out
of sync over time
... woff processor, not defined
ChrisL: agree, if it means UA
jdaggett: no need to define for other than a user agent?
ChrisL: not authoring tools?
Vlad: encoder and decoder need to understad the same format
jdaggett: should not say UA one place and woff processor other places
Vlad: in general agree, but slightly broader scope. Used by tool creators for creation and consumption
<John> Could we have a general term such as 'WOFF client'
ChrisL: as an example a woff validator could say if the checksums are wrong, even though a ua is not required to check them
sylvaing: need to clearly distinguish encoders and decoders. processor could apply to both
jdaggett: agree that sizes should be specified in bytes for example. trivial to add
ChrisL: should say that WOFF stands for web open fornt format
Vlad: another comment from Eric Muller partly addressed by jdaggett's edits
ChrisL: all of Bobs concerns seemed to be addressed by John
Vlad: there was a concern about padding that follows the last table so is not between tables
<Vlad> http://lists.w3.org/Archives/Public/www-font/2010JulSep/0030.html
jdaggett: current spec does not
completely address it
... under overal lfile structure in section 2, (quotes) no
extraneous data outside tha table. its too vague.
... proposed wording, bob replied it would allow data after the
last table
... make change to disallow extraneous data outside the
font
... should UAs load a font that has gaps that are more than
padding
Vlad: think the spec already says this
jdaggett: should be in
conformance requirements
... change 'between font tables' to include before and after
tables too
<scribe> ACTION: jonothan t change conformance rwquirements to disallow extraneous data anywhere and require ua to reject a font contasining it [recorded in http://www.w3.org/2010/08/04-webfonts-minutes.html#action05]
<trackbot> Sorry, couldn't find user - jonothan
<scribe> ACTION: jonathan t change conformance rwquirements to disallow extraneous data anywhere and require ua to reject a font contasining it [recorded in http://www.w3.org/2010/08/04-webfonts-minutes.html#action06]
<trackbot> Created ACTION-13 - T change conformance rwquirements to disallow extraneous data anywhere and require ua to reject a font contasining it [on Jonathan Kew - due 2010-08-11].
Vlad: Eric Muller comments I responded to
ChrisL: surprised by Eric's comments
jdaggett: he seemed to say that everything should be in OT tables (or an extension)
cslye: he said that metadata that
is not woff specific should be inside the fotnt; talked with
him about that
... in private discussion
Vlad: wanted to respond to his comments right away
Vlad: its in 2 weeks
... john suggested woff presentations better done by tal and
erik
... program lists john as the speaker there
erik: agreed
Vlad: please draft something for the group to discuss
erik: will post over the next week
Vlad: important to get the
message across, what woff does and what it is for
... also ChrisL can you do the wg part?
ChrisL: yes sure
... need to tell Tamye about that and check its ok?
cslye: no its ok, the panel discussion is more tightly managed. wg slot is fairly open
ChrisL: so who is on the panel?
cslye: panel is brian mason, cslyye (reads from program)
<cslye> TypeCon webfonts panel discussion: Participants include Bryan Mason, moderator (Typekit), Christopher Slye (Adobe), Bill Davis (Ascender), Tom Phinney (Extensis), Adam Twardoch (FontLab), Raph Levien (Google), Erik van Blokland (LettError), Simon Daniels (Microsoft), Vladimir Levantovsky (Monotype Imaging), John Hudson (Tiro Typeworks), Tal Leming (Type Supply), and more.
John: make sense for ChrisL to be on panel for w3c questions
http://www.typecon.com/program.php
sylvaing: session is on friday? and wg meeting tues?
John: yes and css3 on weds
Vlad: be ready to discuss
conformance requirements and testable assertions
... will get more details on location. 10-6pm is the time. will
get room details
(discussion about publicity)
ChrisL: related my experience on defining 'lossless; wrt PNG spec, might apply to WOFF too
Vlad: fringe cases may need some adjustments
erik: doing experiments with xslt to render the metadata
Vlad: may be useful as informative appendix. jkew said not in spec but can be handy to make available
ChrisL: its more
illustrative
... erik please post a link
erik: sure
vlad: anything else?
adjourned
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) Succeeded: s/jd/jdaggett/ Succeeded: s/Tamaye/Tamye/ Found Scribe: ChrisL Inferring ScribeNick: ChrisL Default Present: Vlad, sylvaing, ChrisL, +31.18.768.aaaa, Erik, tal, John, Sergei, jdaggett, +1.416.589.aabb, Dave, cslye Present: Vlad sylvaing ChrisL +31.18.768.aaaa Erik tal John Sergei jdaggett +1.416.589.aabb Dave cslye Regrets: Christopher Agenda: http://lists.w3.org/Archives/Public/public-webfonts-wg/2010Aug/0010.html Found Date: 04 Aug 2010 Guessing minutes URL: http://www.w3.org/2010/08/04-webfonts-minutes.html People with action items: change conformance jjonathan jjonothan jonathan jonothan rwquirements t[End of scribe.perl diagnostic output]