| Author: | Carsten Haese |
|---|---|
| Contact: | carsten@xiph.org |
| Date: | June 14, 2003 |
The monthly meeting for June 2003 took place on June 8, 2003, at midnight UTC in the IRC channel #xiphmeet on irc.xiph.org. The following people were present, in alphabetical order:
The meeting was moderated by Jack Moffit.
Jack started off by giving a brief update on Xiph.org's organizational status. Our accountant Emily Katt is working on our taxes, and she should have a clear picture how consulting revenue from contract work figures into the taxes.
Also, Xiph's board of directors will meet on June 15. The board meeting will be mostly about financial and legal matters. Minutes will be made available if possible.
Monty is finishing up the Neurosetta project, but there was another last-minute surprise. The compiler that he has to work with is not doing a good job, so he is finding himself writing a lot of assembly by hand, which is extra work and delays the project. The decoder is feature-complete and functional, but it is not working in real time yet.
On the Positron side, things look better. Stan has made 2 beta releases, and most bugs have been fixed. The remaining can either not be reproduced or are kernel-related. Once uploading of Ogg Vorbis files is supported, Positron will be ready for a 1.0 release.
Ralph introduced the Theora development team, which consists of Dan, Mauricio, Timothy, and himself. The alpha 2 release is imminent, and the bitstream format has changed since alpha 1 to accomodate comment and codebook headers, a colorspace indicator, and a cropping rectangle for encoding arbitrary frame sizes.
Jack is working on building binaries of the RealPlayer Vorbis plugin so that users can test the plugin without having to build it. File based playback is working, streaming is not working yet.
Mike informed us about the current status of Icecast. Icecast itself is still in need of autotools work, which Jack is possibly going to look into. Ed has recently revamped the web administration interface. The new interface works well, but is in need of stylistic improvements.
Mike also mentioned an announcement on the icecast mailing list by a group of French students that have done some independent work on icecast. Among technical changes of as-of-yet undetermined merit, they have worked on documentation, which should definitely be integrated into the Icecast project.
Brendan is pushing for a release of libshout, which is mostly complete, except for some testing and documentation.
Dan joined the meeting in progress and gave us a status update on his side of Theora development. Most recently, he was working on restructuring the format of the codebook headers to a compressed Huffman tree representation, which was merged into the main code by Mauricio and Timothy.
Dan's other project is the Theora bitstream specification, which is still in its infancy. He hopes to get it to a point where others can help out and contribute by the end of June.
Regarding the release schedule and future bitstream changes, the general consensus was to release alpha 2 with the current bitstream format as soon as possible, and if further format changes are necessary, an alpha 3 release would incorporate those changes before the format is frozen for the beta 1 release.
Manuel introduced Auratus, Xiph.org's repository for music where artists can sign up and upload music that they wish to release under open licenses such as the Open Audio License or some Creative Commons licenses. Repeater is an Icecast server that will eventually stream tracks from Auratus.
The web programming for Auratus is done, but there are some legal issues that need to be clarified before Auratus can go live. The procedure for how artists will sign up for Auratus is not entirely clear, and the plan is to discuss this with the lawyer at the meeting of the board of directors.
Silvia reminded us that the Ogg encapsulation format and its corresponding MIME type have reached RFC status, and that the story has even made Slashdot. She received some feedback about the RFC, which she forwarded on to the vorbis-dev mailing list.
Phil Kerr reported that he sent an updated draft of the Vorbis-over-RTP specification to the vorbis-dev mailing list. There are only a few open areas left, and he hopes to move it to AVT WG status before the next meeting. He also has begun work on the Internet Draft for the Vorbis bitstream format, which should essentially become an RFC version of the Vorbis specification. And, finally, Phil agreed to take over the Speex-RTP draft.
The fact that the RFC efforts are moving full steam ahead prompted a brief discussion about how Xiph.org's specifications and the RFC's will coexist. The concensus was that the Xiph.org specifications will be the official specifications, and they will be published through the IETF in RFC form.
It was also decided that it's best not to push for IETF standards track, because Xiph.org would have to give up control of the format specifications, and the IETF is more concerned with protocols than formats anyway.
Jack reported that he was contacted by the IEEE-ISTO, which provided a follow-up to Dan's suggestion from last month's meeting that Xiph.org could get funding by becoming a standards body in its own right. The idea is basically that a Xiph Forum would be founded, and interested companies would pay a certain sum to become forum members. The amount of influence and control that the members would get is still open for discussion, but it was agreed by the Xiph board members and the meeting participants that this is a good idea in general.
Jack will contact several companies that might be interested, like e.g. RealNetworks, BBC, and larger game companies, to discuss what kind of involvement these companies would expect in such a forum. Nathan pointed out that the most important requirement for this forum would be to keep a strict Royalty Free policy.
In the last segment of the meeting, Stan presented the work of his sister Sheryl, who volunteered to make up some logos for the individual codecs. The logos were designed to reflect the style of the beloved Xiph logo fish, and the collage of initial attempts can be found at http://www.public.asu.edu/~fuchsia/logo.jpg.
[Note: Current versions of the logos can be found at http://www.rightbrainemporium.com/artwork/artpages/xiph.html.]
Before the meeting was adjourned, an informal discussion about the logos took place. There were many suggestions for improvements, but the overall reaction was very positive.
