See also: IRC log
<trackbot> Date: 10 November 2010
<scribe> scribe: ChrisL
<cslye> "Working Online For Fun"?
<jdaggett> kinda catchy no?
<erik> Isn't that supposed to read "Working Online For Free" ?
Vlad: some good discussion online recently
Vlad: publication date is Nov 16,
seems ok, ends before holiday season
... any problem withthat date?
<cslye> Yes, thanks Chris for all your scribing.
Vlad: comments from Tal on
... comments from David Singer http://lists.w3.org/Archives/Public/public-webfonts-wg/2010Nov/0013.html
ChrisL: makes sense to pad aftyer last table if something follows
jfkthame: think the code for
sfnt2woff leaves meta and private unpadded, but padds all font
... need to check code to be sure
Vlad: failing on lack of padding
at end is not needed
... for the last data block
tal: advantage is that ist clear,
OT is ambiguous and this led to problems
... validator requires padding
ChrisL: also for uncompressed tables, can be used directly
Vlad: need to review text, currently its unclear or contradictory
tal: don't weant to break existing fonts
<erik> (that was tal)
jfkthame: tool leaves meta and
private unpadded if they are the last in the file
... font tables always padded
... metapadded if private follows
... spec and tool always require padding of all font tables including the last one
Vlad: so we should, in section 6,
modify text of para 4 to say that if meta is followd by
private, meta must be padded to four byte boundary
... unless its already on a boundary
<jfkthame> see http://www.w3.org/TR/2010/WD-WOFF-20100727/#Metadata for older wording
jfkthame: WD from July said meta is not padded unless followed by private
cslye: can we just say all tables start on 4 byte boundary and pad to make that so
Vlad: looking for minimal spec
change for clarity
... if meta is followed by private
cslye: we can leave out language on 'if followed'
ChrisL: can we say 0..3 bytes of padding?
Vlad: we can say no additional
padding is needed
... if private is present it must begin on the nearest four byte boundary
... after private no padding is necessary
erik: no need for padding if meta is the last block
Vlad: must begin on a four byte boundary with 0..3 null bytes as padding
<scribe> ACTION: jfkthame to fix up the padding wording per 10 Nov telcon [recorded in http://www.w3.org/2010/11/10-webfonts-minutes.html#action01]
<trackbot> Created ACTION-41 - Fix up the padding wording per 10 Nov telcon [on Jonathan Kew - due 2010-11-17].
jfkthame: is trailing padding allowed?
Vlad: change must to should, padding is ok if tool adds it
sergey: easier for tools to always add padding
Vlad: agree ther eis no need to reject if the trailing padding is absent
<erik> (sergey, not erik)
cslye: in private, there is good language about up to three bytes, can we reuse that?
Vlad: font tables always padded so meta always syarts ona four byte boundary
jfkthame: ok, that works
Vlad: that deals with one of dave
singers comments and one of mine
... chris suggested language for what 'valid metadata' means
<scribe> ACTION: jfkthame to update defination of valid metadata [recorded in http://www.w3.org/2010/11/10-webfonts-minutes.html#action02]
<trackbot> Created ACTION-42 - Update defination of valid metadata [on Jonathan Kew - due 2010-11-17].
'valid metadata is well formed, matches the schema and is compressed'
jfkthame: why is it needed?
Vlad: concern was whether
rejection on uncompressed or not well formed was neeeded
... or avoid the term valid
... now on the metadata
... uniqueid element is optional but if present, the attribute should be required
ChrisL: spec does not say that explicitly, the schema does require it
Vlad: spec says that in a discrepancy between spec and schema, the text takes precedence. so the text needs to be clear here
<scribe> ACTION: jfkthame to clarify that @id on uniqueid is required [recorded in http://www.w3.org/2010/11/10-webfonts-minutes.html#action03]
<trackbot> Created ACTION-43 - Clarify that @id on uniqueid is required [on Jonathan Kew - due 2010-11-17].
Vlad: next one is a typo, credits element must contain at least one credit element, the must is lowwrcase, should be MUST
jfkthame: all of the elements say that they must be a child of the metadata element, maybe all of these should be MUST
Vlad: language on text subelements says they are localisable but does not say that you need at least one of them
<scribe> ACTION: jfkthame to clarify schema language in text to say one or more text elements [recorded in http://www.w3.org/2010/11/10-webfonts-minutes.html#action04]
<trackbot> Created ACTION-44 - Clarify schema language in text to say one or more text elements [on Jonathan Kew - due 2010-11-17].
jfkthame: but they can still make empty text elements
Vlad: item element as part of
extyension element, current text says at least one should.
should be a MUST
... otherwise you can have an empty extension element
jfkthame: what is the harm?
(we go with MUST)
<scribe> ACTION: jfkthame to make item element a required chilkd of extension [recorded in http://www.w3.org/2010/11/10-webfonts-minutes.html#action05]
<trackbot> Created ACTION-45 - Make item element a required chilkd of extension [on Jonathan Kew - due 2010-11-17].
Vlad: couple of missed testable assertions
action chris to add the two missing conformance statements and split out the stylesheet
<trackbot> Created ACTION-46 - Add the two missing conformance statements and split out the stylesheet [on Chris Lilley - due 2010-11-17].
Vlad: some minor changes from jdaggett as well
(agree with jdaggett changes)
ChrisL: feedback was that getting a font/ top level type was a lot of lengthy work
Vlad: nice to have but risk of
... if that is the consensus we could go straight with application/font-woff
ChrisL: this could easily consume a year or two . hard to justify
resolved: go with application/font-woff
<scribe> ACTION: chris to edit the media registration to say application/font-woff [recorded in http://www.w3.org/2010/11/10-webfonts-minutes.html#action06]
<trackbot> Created ACTION-47 - Edit the media registration to say application/font-woff [on Chris Lilley - due 2010-11-17].
<trackbot> ACTION-23 -- Jonathan Kew to verify whether, with zlib, it's possible to have a stream that allows an EOF midstream, so extra data can be padded on mid-stream. -- due 2010-08-24 -- CLOSED
<trackbot> ACTION-33 -- Chris Lilley to write up the proposed text for WOFF processing model -- due 2010-10-27 -- OPEN
<trackbot> ACTION-33 Write up the proposed text for WOFF processing model closed
<trackbot> ACTION-34 Add id and class to testable assertions closed
<trackbot> ACTION-39 Write to iesg suggesting font/ top level type closed
<trackbot> ACTION-40 Add a media type registration remplate for woff, omitting the top level type closed
Vlad: any objections?
resolved: request last call for woff specification
Vlad: timeframe for spec edits?
jfkthame: will do tomorrow
Vlad: please send email to list once spec edits are done. we will then consider it ready for publication
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/erik: don't/tal: don't/ Succeeded: s/erik: easier/sergey: easier/ Succeeded: s/NUST/MUST/ Found Scribe: ChrisL Inferring ScribeNick: ChrisL Default Present: ChrisL, Vlad, +44.845.397.aabb, +1.443.895.aacc, +1.425.213.aadd, jfkthame, +1.408.536.aaee, cslye, +31.70.360.aaff, Erik, tal Present: ChrisL Vlad +44.845.397.aabb +1.443.895.aacc +1.425.213.aadd jfkthame +1.408.536.aaee cslye +31.70.360.aaff Erik tal Regrets: dave_crossland john_hudson Agenda: http://lists.w3.org/Archives/Public/public-webfonts-wg/2010Nov/0017.html Found Date: 10 Nov 2010 Guessing minutes URL: http://www.w3.org/2010/11/10-webfonts-minutes.html People with action items: chris jfkthame[End of scribe.perl diagnostic output]