RE: Telcon minutes, Wed. Jun 12th

Hello WG,

Since all of the compression results from major web font collections are now in [1], and looking at the breakdown of the results that shows the impact of each component of the proposed solution [2] - it can be clearly seen that while the pre-processing step on average consistently produces better gains (i.e. higher averages with lower StDev) - the LZMA is a critical component that is responsible for all *stellar* results where max compression gains far exceed what gzip or similar technology can offer. It is clear to me (and I hope that you share the same opinion) that implementing entropy coding in WOFF 2.0 using LZMA compression is a definite necessity, even though it is one step that will most likely require significant time and effort to accomplish, especially taking into consideration the needs to document it and produce CTS suite with sufficient test coverage. It is a big undertaking and I want us all to realize it clearly and also consider having additional experts involved to help us accomplish this goal.

Our most recent telcon discussion was focused mostly on the LZMA outlining the sentiment mentioned above and also emphasizing the need for LZMA implementation to be security tested - something that will require involvement and help from not only font experts in the group but also security and compression experts. I would like to ask all WG members to discuss this within your respective organizations and invite people who can help us to accomplish our goals to join the WG. If you know of anyone who is independent and would be willing to serve as an Invited Expert on the WG - please let us know as well.

Thank you,
Vladimir


[1] http://lists.w3.org/Archives/Public/public-webfonts-wg/2013Jun/0006.html
[2] http://lists.w3.org/Archives/Public/public-webfonts-wg/2013Jun/0007.html



From: Levantovsky, Vladimir [mailto:Vladimir.Levantovsky@monotype.com]
Sent: Wednesday, June 12, 2013 11:22 AM
To: public-webfonts-wg@w3.org
Subject: Telcon minutes, Wed. Jun 12th

http://www.w3.org/2013/06/12-webfonts-minutes.html

and as plain text below:
                               - DRAFT -

                 WebFonts Working Group Teleconference

12 Jun 2013

   See also: [2]IRC log

      [2] http://www.w3.org/2013/06/12-webfonts-irc

Attendees

   Present
   Regrets
   Chair
          SV_MEETING_CHAIR

   Scribe
          Vlad

Contents

     * [3]Topics
     * [4]Summary of Action Items
     __________________________________________________________

   <trackbot> Date: 12 June 2013

   Sergei, are you going to call in as well?

   Sorry, I just saw your msg to the IRC

   For the records - zakim is deaf and blind (doesn't see anyone
   on the phone)

   On the phone - we have David, Jonathan and Vlad

   [5]http://lists.w3.org/Archives/Public/www-font/2012JanMar/0002
   .html

      [5] http://lists.w3.org/Archives/Public/www-font/2012JanMar/0002.html

   For the IRC only participants - Discussing compression gain
   results and relative contributions that can be seen from
   preprocessing steps and use of LZMA

   JK - talking about risks introducing LZMA and potential
   vulnerability from that

   Sergei - have you had MS security team look at the LZMA code to
   see if this is usable for you?

   <sergeym> Sorry, I went offline for a few minutes

   Or rather, do you know if Sylvain had them to look at LZMA

   <sergeym> No, Sylvain didn't

   <sergeym> Matt Rakow is now taking over this work

   <MaRakow> hi :)

   taking over security analysis or webfonts in general?

   Hi Matt

   <MaRakow> TBD - sorry, I'm still quite new in this space

   discussing LZMA use and security concerns to be addressed

   still on the phone - discussed Monotype WFS compression results
   and now talking again about allocating some resources from each
   member to address LZMA concerns

   On the IRC were Jonathan, David, Sergei, Matt, Vlad

Summary of Action Items

   [End of minutes]
     __________________________________________________________