Re: More information on MAS
- From: Janina Sajka <janina rednote net>
- To: Bill Haneman <bill haneman sun com>
- Cc: gnome-multimedia gnome org
- Subject: Re: More information on MAS
- Date: Thu, 6 Feb 2003 10:48:13 -0500
Good list of why we need synchronization in media streams, Bill.
May I also add the DAISY Digital Talking Book as another? Audio and/or text titles are already in distribution to blind patrons through library services
in Sweden, the U.K., Japan, the U.S., and elsewhere. The next iteration of these SMIL based specs will likely become the basis of curricular materials
distribution for print disabled studends (K-12) in the U.S. In fact, ACB, AFB, NFB and the Association of American Publishers have come to agreement on
a legislated solution that was proposed in our Congress last year. The legislation, known as the Instructional Materials Accessibility Act (IMAA) has
again been introduced in the current Congress and can be tracked, day to day, at:
http://www.afb.org/textbooks.asp
One result of last year's attempt to pass this law is the formation of a study group by our Department of Education and chaired by CAST:
http://www.cast.org/ncac/index.cfm?i=3138
In fact there's tremendous opportunity here for leadership in the development of good software tools. Some key points:
* The basis of the Digital Talking Book is open file and markup specifications from the W3C, particularly XML and SMIL. Current deployments use
DAISY 2.02, and the next iteration of these specifications has already been completed and approved as ANSI Z39.86-2002:
http://www.daisy.org/publications/specifications.asp
The sole exception is the lack of an open audio compression protocol in the ANSI spec--but there is movement toward adopting OGG-Vorbis, or
perhaps more appropriately Speex:
http://www.speex.org
* All user agents and authorinig tools currently available to read or produce DAISY content are proprietary, though there is interest in the
DAISY Consortium to move into open source particularly with their need to support dissemination of this powerful technology to emerging nations. There
is, in fact.
* It is already evident that the DAISY technology is attracting main stream publishing interest. The first U.S. Government document to use DAISY
2.02 was released last spring. The first commercial audiobook title will be released in March by AOL/Time-Warner Audio Books.
* The mainstream publishing industry is short on integrated authoring tools for producing any kind of multimedia content, let alone DAISY. Most
publishers are still using Quark, and they don't have nice things to say about it. If you know Quark and how it's used in publishing, you know that it's
focused squarely on paper printing.
I raise all of this because most often conversations about "multimedia" don't think in terms of books. That's a mistake, and I think a big mistake. I
will stop for now. But were I to continue I could develop arguments suggesting that non-entertainment is a bigger commercial opportunity for multimedia
publishing. Some ideas about why are in the paper I coauthored with George Kerscher entitled "Surpassing Gutenberg:"
http://www.afb.org/ebook.asp
Bill Haneman writes:
> From: Bill Haneman <bill haneman sun com>
>
>
> >
> > Mike Andrews posted this to Slashdot with some information about the
> > rationale behind MAS.
> > http://developers.slashdot.org/comments.pl?sid=52708&cid=5220114
> >
> > They haven't managed to fix their webpage yet with the 'only works on
> > high resoltions that I mailed Mike about, but I guess you can't win em
> > all ;)
> >
> > Christian
> >
> > Also as a sidenote, Zeeshan Ali has already begun working on a set of
> > MAS source and sinks for GStreamer, the first code for this is already
> > checked into the head branch, but if I understood Zeeshan correctly he
> > needs some more info from Mike to make it work right.
>
> One of the main benefits of MAS is its ability to support
> synchronization. This is required before we can do anything with SMIL
> (which Janina Sajika recently enquired about). In order to support good
> quality synchronization of multiple media streams, you need good latency
> information, completion notifications, etc. and my understanding is that
> this is something MAS was designed to handle and which is
> difficult/impossible to get right with out other media sinks.
>
> A general example is synchronized audio and video; particularly for
> codecs that don't support interleaving of audio and video. And most of
> our multimedia formats don't have any kind of sideband or side-channel
> for captioning, so captioning, descriptive audio, subtitles, etc. can't
> reliably be synchronized with the 'primary' AV stream. This can be a
> problem even on a desktop machine, but it really gets nasty if you are
> doing anything with streamed content or remote AV applications, etc.
> It's even a problem for something as apparently-simple as scrolling
> captions and braille output, since the text stream needs to pause and
> resync if the audio stream pauses, glitches, or just has latency issues.
>
> regards,
>
> Bill
>
>
> --
> Bill Haneman <bill haneman sun com>
>
> _______________________________________________
> gnome-multimedia mailing list
> gnome-multimedia gnome org
> http://mail.gnome.org/mailman/listinfo/gnome-multimedia
--
Janina Sajka, Director
Technology Research and Development
Governmental Relations Group
American Foundation for the Blind (AFB)
Email: janina afb net Phone: (202) 408-8175
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]