<trackbot> Date: 01 September 2010

<John> Regret: not able to phone in today. Will try to follow IRC for a while and comment as appropriate.

going through Vlad's draft presentation for ATypI for any comments/suggestions

I have the email with attachment

<Vlad> discussed the slides sent via email, no changes - please submit your comments over email

Jonathan proposes to drop the requirement to verify the checksums.

Sylvain: is there data that shows fonts will fail on one platform but not on another based on faulty checksums?

Jonathan: WOFF is a container. Why single out one detail of the OFF format and make it a conformance item?

Vlad: Firefox checked the checksums?

Jonathan; any structurally sound SFNT file should be able to be packed. Regardless of the contents.

Vlad: checksum helps to establish the file is sound

Sergey: other structural checks are essential for making sure we can be lossless
... validating checksums is one of the steps

Sylvain: I don't see why

Sergey: woff creation tool should validate sfnt structure
... validating checksum is part of that

Christopher: do we want woff creation tools to validate that it is a valid sfnt?

Vlad: spec mentions "sfnt" many times

<cslye> Are there security concerns? Is that a reason to ensure a valid sfnt?

<scribe is struggling a bit>

vlad: we have to make sure the input is structurally sound

sergey: if UA does partial decompression, it can't use the checksums

jonathan: define - what is a valid woff file?

vlad: recap: agreement on checking the input file on structural integrity
... offsets proper values, and so on
... still split on whether checksums are part of the structure or part of the semantics

julio: if we go down this path / some fonts developed for corporate use might not work
... there may be inoffensive problems in a font

vlad: if a font fails, user should contact the maker and have it fixed

julio: maker might not be available

sergey; spec offers a choice: reject or make it work

julio: worried specifically about corporate fonts

vlad: checking is within scope; which actions to take it not in scope

<jfkthame> so if a woff file contains incorrect table checksums, is it a valid file or not?

vlad, out of time

<Julio> I have to check-out - thanks guys

jonathan: still fuzzy about the definition of what is a valid woff file
... does the definition include the checksums?

sergey: we can say a valid file has a valid checksum. if checksum is incorrect: UA behaviour is undefined

<Vlad> Meeting: WebFonts WG

<Vlad> Scribe: Erik

John had to leave early

<Vlad> ScribeNick: erik

christopher: end the meeting and continue online?

vlad: next week I will be at atypi
... volunteers for replacing the chair, or postponing the call

christopher: we could use the list better - better documented in email

http://www.w3.org/Fonts/WOFF-FAQ

http://typophile.com/node/73465

wrapping up.

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)

Found Scribe: Erik
Inferring ScribeNick: erik
Found ScribeNick: erik

WARNING: No "Present: ... " found!
Possibly Present: John Jonathan Julio Regret ScribeNick Sergey Sylvain Vlad christopher cslye jfkthame joined sergeym trackbot webfonts
You can indicate people for the Present list like this:
        <dbooth> Present: dbooth jonathan mary
        <dbooth> Present+ amy

Agenda: http://lists.w3.org/Archives/Public/public-webfonts-wg/2010Aug/0068.html
Found Date: 01 Sep 2010
Guessing minutes URL: http://www.w3.org/2010/09/01-webfonts-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.