CARVIEW |
- DRAFT -
Efficient XML Interchange Working Group Teleconference
13 Dec 2016
See also: IRC log
Attendees
- Present
- Regrets
- Chair
- SV_MEETING_CHAIR
- Scribe
- taki
Contents
CB: I talked to CSS team contact,
who told me it will be on the telecon agenda.
... I will need to ping him again.
Empty DTR Map same as no DTR Map?
DP: EXI stream that use DTRM. in
a way there is no types. Empty DTRM element in the
header.
... Exificient parsed it correctly. The assumption is that
there is no DTRM. Codec fallbacks to the default.
... The encoder expected differently.
... In Canonical EXI as well we say, if DTRM is empty, we can
remove it.
... I am not sure spec is clear enough.
TK: We need a testcase in TTFMS.
application/exi media-type
DP: It seems that application/exi media-type is missing from IANA.
CB: I will contact liaison contact.
EXI's built-in DEFLATE compression still the best way to go?
DP: I further tested. ZSTD from
facebook and LZMA
... Both don't seem to beat Brotli.
... At least in default mode.
<caribou> https://www.w3.org/TR/WOFF2/ uses Brotli compression
DP: I looked into CSS.
... I tried some optimizations. I essentially changed the
schema so that well-known CSS property names are
enumerated.
... I tried to compare the result.
... This difference occurs if you can use strict and enumerated
mode.
... We definitely need grammar strings.
... So that the structure is strict but content can be
extended.
... I added ZSTD and LZMA in excel.
... I may be possible that those algorithms may work better
with EXI.
... Once you use good algorithm like brotli, even EXI
bytealigned almost works as good as precompression.
CB: WOFF2 requires brotli, and only brotli.
<caribou> https://www.w3.org/2004/01/pp-impl/44556/show-license#x35662
<caribou> https://datatracker.ietf.org/doc/rfc7932/
<caribou> 'informational'
<caribou> https://datatracker.ietf.org/doc/review-alakuijala-brotli-09-genart-telechat-kyzivat-2016-04-28/ (oops....)
<caribou> IPR is clearly covered (https://datatracker.ietf.org/ipr/2396/)
Efficient XML Interchange (EXI) Extended String
DP: I was wondering, if it is
useful to have a flag to assert there is grammar string is
specified.
... For simplicity and performance.
... At "appinfo" level. To indicate the appinfo is for EXI.
TK: I will add it in section 3.
Our next telecon will be on January 10.
Summary of Action Items
Summary of Resolutions
[End of minutes]Minutes formatted by David Booth's scribe.perl version 1.148 (CVS log)
$Date: 2016/12/13 20:03:19 $
Scribe.perl diagnostic output
[Delete this section before finalizing the minutes.]This is scribe.perl Revision: 1.148 of Date: 2016/10/11 12:55:14 Check for newer version at https://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) No ScribeNick specified. Guessing ScribeNick: taki Inferring Scribes: taki WARNING: No "Present: ... " found! Possibly Present: CB DP TK caribou exi https joined trackbot You can indicate people for the Present list like this: <dbooth> Present: dbooth jonathan mary <dbooth> Present+ amy <amy> Present+ WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 13 Dec 2016 Guessing minutes URL: https://www.w3.org/2016/12/13-exi-minutes.html People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]