WebFonts Working Group Teleconference -- 08 Jan 2014

<trackbot> Date: 08 January 2014

<ChrisL> vlad, updated evaluation report at http://www.w3.org/Fonts/WG/WOFF2ER/

<Vlad> Thanks, Chris

<ChrisL> scribenick: fil

<kuettel> I haven't been able to join the conference call just yet. Trying again

predictive points

discussing email from jonathan

vlad: expected minimal benefit, but didn't expect large increase in size
... From the data can conclude that predictive points aren't useful. It was a great idea and exploration

<ChrisL> I summarised this here http://www.w3.org/Fonts/WG/WOFF2ER/#preproc

<ChrisL> but without the data yet

even though we can get reduction of font input, once it goes through the entropy coder there is a penalty

ChrisL: wants to keep specs terse and easily implementable, and keep data in the appendices

patent

<kuettel> Update from Fil on patents

<kuettel> Vlad giving quick background on the topic

<kuettel> The question that came up before, was what IP policy / path is being pursued for Brotli

<kuettel> Chris, said that the precedent is to use a patent grant

<kuettel> Vlad: is the goal to pursue the RFC track?

<kuettel> Fil: reached out to the open source / patent team internally

<kuettel> Fil: the team suggested that the W3C patent policy could accomplish the goal

<kuettel> Fil: with the RFC track (and precedent for spdy work), there would be a disclosure / grant

SPDY disclosure: https://datatracker.ietf.org/ipr/1692/

<kuettel> Fil: posting an example link

<kuettel> Chris: eager to research this further and confirm that this would be a good path to pursue

<kuettel> Fil: also noting that the code was released under the Apache license

<kuettel> Fil: thus, in using the code it would be a non-issue

<kuettel> Fil: but we would still like to explore the disclosure

<kuettel> Fil: patent team is eager to take the steps necessary to ensure that others can fully implement Brotli / WOFF 2.0 without issue

<kuettel> Vlad: a broader grant (not just limited to W3C use) would be ideal

<kuettel> Fil: let's confirm in the next meeting w/ Chris that this is the right path to pursue

Brotli security

Chrome and Google Fonts have started security testing, initally with fuzz testing

Work has started and it's looking good

ChrisL: Will that work be published anywhere, and what is the extent?

kuettel: Will reach out to Chrome team to ask about what can be released

Goals for next few months

Almost done with evaluation report, so can hopefully publish soon.

If Brotli is published as RFC, W3C group can focus on WOFF pieces

kuettel: To follow up with Raph about the draft progress

vlad: Probably will cancel next week's call
... should ask on email for call for consensus, ask for objections within a week

FPWD

<ChrisL> http://www.w3.org/Fonts/WG/WOFF2ER/

<Vlad> http://www.w3.org/Fonts/WG/WOFF2ER/

<ChrisL> Vlad: call for objections to FPWD, will do call for consensus on mailing list as well

<ChrisL> resolved: publish FPWD pending confirmation i=on mailing list

Next step: need a draft that can be presented to group members

Does it make sense to start working on longer-term projects like conformance tests?

Will want to run the same tests like WOFF 1

Look at all statements in the spec that were testable assertion, and write tests for them

So, there's a need for a stable spec before tests are written.

Vlad: because of departures from WOFF1, won't be able to reuse a lot of the tests

With two-step (preprocessing, entropy coder), might want to test steps separately.

Bitwise comparison might not be accurate -- eg with hint code, might have same functionality without same bits

Vlad: We should focus on spec work

Changes from original MTX submisssion

ChrisL: Can't find where there were pieces of the original MTX that were removed

Vlad thinks that this is in font-compression-reference on google code

<kuettel> https://code.google.com/p/font-compression-reference/source/browse/docs/WOFFUltraCondensedfileformat.pdf

<ChrisL> great, thanks

actions cleanup

<Vlad> http://www.w3.org/Fonts/WG/track/

<kuettel> Actually, here is the right link for the font-compression-reference document:

<kuettel> http://wiki.font-compression-reference.googlecode.com/git/img/WOFFUltraCondensed.pdf

<ChrisL> close action-105

<trackbot> Closed action-105.

<ChrisL> close action-110

<trackbot> Closed action-110.

<ChrisL> close action-111

<trackbot> Closed action-111.

<ChrisL> close action-112

<trackbot> Closed action-112.

<ChrisL> action-115?

<trackbot> action-115 -- Chris Lilley to And Vlad to put together the draft for top-level font registration, possibly based on text from previous attempts. -- due 2012-11-08 -- OPEN

<trackbot> http://www.w3.org/Fonts/WG/track/actions/115

<ChrisL> related to action-121

<ChrisL> Vlad: need to document why it is needed

<ChrisL> Vlad: have earlier drafts as a basis, its the rationale that is crucial here

<ChrisL> close action-115

<trackbot> Closed action-115.

<ChrisL> close action-119

<trackbot> Closed action-119.

<ChrisL> we can also close 123 and 124 as those were done

<ChrisL> close action-117

<trackbot> Closed action-117.

<ChrisL> done, tools open source

<ChrisL> close action-123

<trackbot> Closed action-123.

<ChrisL> close action-124

<trackbot> Closed action-124.

<ChrisL> ACTION: chris to reach out on patent for brotli [recorded in http://www.w3.org/2014/01/08-webfonts-minutes.html#action01]

<trackbot> Created ACTION-125 - Reach out on patent for brotli [on Chris Lilley - due 2014-01-15].

<ChrisL> ACTION: david to check publication of security review [recorded in http://www.w3.org/2014/01/08-webfonts-minutes.html#action02]

<trackbot> Created ACTION-126 - Check publication of security review [on David Kuettel - due 2014-01-15].

ChrisL: Has Brotli been used for other data types, like SVG?

Kuettel: Doesn't think so, mostly been compressing text with brotli.

MPEG meeting next week, might want to try compressing a few examples before.

Vlad: might get an indication of compression by compressing xml.

ChrisL: Svg has a lot of path data, which was designed to be working well with gzip.

<scribe> ACTION: kuettel to work with ChrisL to try compressing a few svgs with Brotli [recorded in http://www.w3.org/2014/01/08-webfonts-minutes.html#action03]

<trackbot> Created ACTION-127 - Work with chrisl to try compressing a few svgs with brotli [on David Kuettel - due 2014-01-15].

<ChrisL> http://people.mozilla.org/~jkew/opentype-svg/GeckoEmoji.ttf

<ChrisL> http://people.mozilla.org/~jkew/opentype-svg/GeckoEmoji.woff

<ChrisL> http://people.mozilla.org/~jkew/opentype-svg/soccer.otf

<ChrisL> http://www.w3.org/2013/10/SVG_in_OpenType/

Vlad: has there been anything done before to preprocess XML for compression?

<ChrisL> ACTION: chris to report on svg binaryxml as a preprocessing step [recorded in http://www.w3.org/2014/01/08-webfonts-minutes.html#action04]

<trackbot> Created ACTION-128 - Report on svg binaryxml as a preprocessing step [on Chris Lilley - due 2014-01-15].

<ChrisL> (adjourned)

<ChrisL> sigh

<ChrisL> chair: vlad

<ChrisL> Scribe: fil

- DRAFT -

WebFonts Working Group Teleconference

08 Jan 2014

Attendees

Contents