19:57:03 <RRSAgent> RRSAgent has joined #webfonts
19:57:03 <RRSAgent> logging to http://www.w3.org/2014/03/26-webfonts-irc
19:57:05 <trackbot> RRSAgent, make logs world
19:57:05 <Zakim> Zakim has joined #webfonts
19:57:07 <trackbot> Zakim, this will be 3668
19:57:07 <Zakim> ok, trackbot; I see IA_Fonts()4:00PM scheduled to start in 3 minutes
19:57:08 <trackbot> Meeting: WebFonts Working Group Teleconference
19:57:08 <trackbot> Date: 26 March 2014
19:57:45 <Zakim> IA_Fonts()4:00PM has now started
19:57:52 <Zakim> +Vlad
20:00:00 <Zakim> +[Microsoft]
20:00:23 <sergeym> sergeym has joined #webfonts
20:00:24 <Zakim> +jfkthame
20:00:39 <jfkthame> jfkthame has joined #webfonts
20:02:00 <Zakim> +[Google]
20:02:03 <raph> raph has joined #webfonts
20:07:52 <Zakim> +[Google.a]
20:08:09 <kuettel> kuettel has joined #webfonts
20:12:18 <raph_> raph_ has joined #webfonts
20:12:36 <raph_> first agenda item: applause
20:12:47 <Vlad> scribenick: rapph_
20:12:56 <raph_> scribenick: raph_
20:13:08 <raph_> going from discussion to first editor's draft is major progress
20:13:26 <raph_> now that we have it written down, there'll be plenty to talk about
20:13:31 <raph_> who has reviewed the spec?
20:14:02 <raph_> kuettel, jfkthame, and sergeym have done cursory review
20:14:36 <raph_> vlad has done a full pass over the spec, and has a number of comments, both editorial and deeper
20:14:46 <raph_> introduction should have reference to woff 2 evaluation report
20:14:58 <raph_> datatypes
20:15:17 <raph_> should, as much as possible, use same datatypes as opentype and off specs
20:15:30 <raph_> mixed used of USHORT and UINT16, should settle on one
20:16:22 <Zakim> + +1.250.668.aaaa
20:16:28 <jhudson> jhudson has joined #webfonts
20:16:29 <raph_> this is a terminology question
20:16:40 <raph_> raph: agree, should use opentype terminology for these types
20:17:03 <raph_> datatype uintbase128
20:17:16 <raph_> what is the motivation for introducing it?
20:17:25 <raph_> used only in table directory
20:17:53 <raph_> is there a real statistical importance that there are many lengths that are <=127
20:18:20 <raph_> in original opentype spec, 4 byte values
20:18:42 <kuettel> raph: one use case, really small subsets
20:18:59 <kuettel> raph: for these fonts, the overhead of the table header is a large %
20:19:27 <kuettel> raph: also see other fonts, where many tables have a small length (e.g. heea)
20:19:57 <kuettel> raph: most tables, except for Glyph, will have a two byte encoding
20:20:28 <kuettel> raph: original proposal had a long and short table format (similar to WOFF 1.0)
20:20:36 <kuettel> raph: difference was typically 200 bytes
20:21:00 <kuettel> vlad: would be good to elaborate on this particular data type
20:21:26 <kuettel> raph: sounds good, should be easy to quantify the size reduction
20:21:59 <kuettel> vlad: going further, paragraph with the flags
20:22:10 <kuettel> vlad: e.g. known table tags
20:22:29 <kuettel> vlad: would we regret this later, esp. as new tables are added (e.g. color fonts)
20:23:40 <kuettel> raph: right, great question
20:24:28 <kuettel> raph: the thinking is, that the list of tables -- we will see more fonts with additional tables
20:24:51 <kuettel> raph: but, the majority of the tables will still be in the OpenType spec
20:25:17 <kuettel> raph: vast majority of tables on the list were from the standard
20:25:38 <kuettel> raph: have to come up with a set (justification is the number of bytes saved)
20:26:41 <kuettel> raph: any new table would get passed through, it just wouldn't benefit from this particular four-byte per table savings
20:27:28 <kuettel> jonathan: asking about the number of bits required to encode...
20:27:44 <kuettel> raph: original motivation, was to support multiple compression formats
20:28:03 <kuettel> raph: proposal is to only support Brotli (and not gzip)
20:29:40 <kuettel> vlad: as a follow up, did not appear as if the last to bits were not for the compression type
20:30:00 <kuettel> raph: yes, compression type proposal (Brotli) only, is encoded in the draft
20:31:31 <kuettel> vlad: both bit allocation for table tags, and bit 5 for pre-processing, identified for future discussions
20:32:18 <kuettel> vlad: wouldn't it be easier to always process all tables
20:33:14 <kuettel> raph: minimizing option is a good goal.  motivations was to allow trade-off between processing time/memory
20:33:35 <kuettel> raph: another would be to preserve byte level data
20:34:10 <raph_> jfkthame: similar tradeoff as in choice of entropy coder
20:34:57 <raph_> vlad: transform is much less computation than entropy coding
20:35:40 <raph_> jfkthame: impact on encode time is one reason people may choose less expensive processing
20:35:41 <kuettel> vlad: bringing discussion back to preprocessing
20:36:14 <kuettel> raph: reconstructing the bounding boxes is not trival
20:36:31 <kuettel> raph: could be good to review the data
20:37:00 <kuettel> raph: use cases for fast encoding, or want to enable fast decode (e.g. for mobile)
20:37:26 <kuettel> raph: complexity of not having the transform (e.g. the bit) is trival
20:38:29 <kuettel> vlad: agree, rather wondering if preprocessing could always be kept
20:39:44 <kuettel> vlad: would like to give the working group more time to review, and then once decided would likely want to elaborate
20:41:18 <raph_> vlad: section 4 should be "compressed data format" rather than "general table format
20:41:46 <raph_> vlad: description of bitmask should be part of glyph header
20:42:57 <raph_> raph: agree, was concerned about the fact that bitmask is variable size, but having it part of header is cleaner
20:43:55 <raph_> vlad: "explicity bitmap" seems like typo
20:44:09 <raph_> raph: yes, should read "explicitly set bounding box" instead
20:45:34 <kuettel> raph: would like to ask working group, does the draft look pretty complete
20:45:56 <raph_> vlad: reasonably complete, we know some pieces are missing (like triplet encoding)
20:45:56 <kuettel> vlad: looks pretty complete, noting the few sections called out in the document that we need to fill in
20:46:20 <raph_> vlad: some more things that need to be discussed
20:46:39 <raph_> will need to go through and make sure all normative statements are properly defined
20:46:51 <raph_> that may happen after public working draft
20:47:46 <kuettel> raph: update from the Compression Team on Brotli.
20:48:05 <kuettel> raph: they consider the format final now, and will next focus on publishing the IETF draft
20:50:10 <raph_> kuettel: how many compression formats should we support?
20:50:17 <raph_> can we standardize on just Brotli?
20:52:08 <raph_> vlad: woff 1 offers no compression and gzip
20:52:26 <raph_> so need to offer these options in woff 2 doesn't make much sense
20:52:45 <raph_> significant improvement in compression efficiency was main motivation for woff 2
20:52:53 <raph_> vlad: votes for "keep it simple"
20:53:14 <raph_> kuettel: compression team provided numbers on compression performance
20:53:32 <raph_> fast brotli encoder gets 10% better compression than gzip at no additional cost in encode time
20:53:40 <raph_> numbers to be pasted into chat
20:53:57 <kuettel> Some numbers from the Compression Team on the Brotli fast version:
20:54:06 <kuettel> woff2/gzip:  60,844,848 bytes, 13.6 Mb/s encoding speed, 104.2 Mb/s decoding speed
20:54:16 <kuettel> woff2/brotli: 54,232,880 bytes, 13.2 Mb/s encoding speed,  84.4 Mb/s decoding speed
20:54:34 <kuettel> And then an update on the slow version:
20:54:43 <kuettel> the decoding speed of woff2 produced with a slow brotli encoder got faster as well, it is now 80.6 Mb/s, while the total corpus size is 50,112,368 bytes.
20:54:53 <raph_> vlad: extra bit freed up for compression type may help expand known tags
20:55:14 <raph_> vlad: we should consider the question of whether to make glyf transform optional (current draft) or required
20:55:34 <raph_> implementation an option to give less than maximal compression seems counterintuitive
20:55:54 <raph_> vlad: doesn't like optional stuff
20:56:16 <raph_> vlad: would rather avoid it
20:56:31 <raph_> jfkthame: that decision needs more time to think through
20:56:55 <raph_> vlad: raising issue, not expecting a decision today
20:57:05 <raph_> jfkthame: another change to wire format to propose
20:57:12 <raph_> flags in table directory entry
20:57:29 <raph_> only thing doing with bits 6 and 7 is noting whether table is first in the font
20:57:50 <raph_> and the only thing that's used for is whether to include compressed stream length
20:58:07 <raph_> would be simpler to move into woff header, and this would also free up an additional bit
20:58:26 <raph_> vlad: what is value of having table directory preserved?
21:00:09 <raph_> vlad: compressed data stream might include table directory
21:01:20 <raph_> in mtx, everything was part of compressed stream, including table directory
21:04:13 <kuettel> Sorry, I have to join another meeting
21:04:17 <Zakim> -[Google.a]
21:04:42 <raph_> jfkthame: woff2 header could also (mostly) be inside
21:04:53 <raph_> raph: metadata could also be inside compressed stream
21:05:07 <raph_> vlad: makes sense for metadata to be separately accessible
21:05:20 <raph_> use cases where you want to read metadata without having to decompress entire font
21:05:56 <raph_> jfkthame: metadata should be brotli-encoded
21:06:09 <raph_> raph: yes, already in spec
21:06:51 <raph_> vlad: will be traveling next week for iso/mpeg meeting, finalizing 3rd edition of OFF spec
21:07:09 <raph_> vlad: let's have more discussion over email
21:07:19 <raph_> then telcon will be more productive
21:07:50 <raph_> next conference call in 2 weeks, have discussion of specific items during that time
21:09:42 <raph_> raph: i hope to check spec into W3C CVS soon
21:10:17 <Zakim> - +1.250.668.aaaa
21:10:26 <Vlad> zakim, list attendees
21:10:26 <Zakim> As of this point the attendees have been Vlad, [Microsoft], jfkthame, [Google], +1.250.668.aaaa
21:10:37 <raph_> end of call
21:10:46 <Vlad> rrsagent, make minutes
21:10:46 <RRSAgent> I have made the request to generate http://www.w3.org/2014/03/26-webfonts-minutes.html Vlad
21:10:56 <Zakim> -[Google]
21:11:00 <Zakim> -Vlad
21:11:02 <Zakim> -jfkthame
21:12:59 <Zakim> -[Microsoft]
21:13:00 <Zakim> IA_Fonts()4:00PM has ended
21:13:00 <Zakim> Attendees were Vlad, [Microsoft], jfkthame, [Google], +1.250.668.aaaa
23:29:23 <Zakim> Zakim has left #webfonts