[Foundation-l] Video accessibility

[Michael Dale]

Siliva is a log time collaborator with the metavid project around 
metadata. I hope to continue to work with her to ensure the wiki 
extension fits well with the proposed accessibility features of firefox.

MetavidWiki and mv_embed is built around supporting cmml the timed text 
format Siliva authored. CMML already has a well defined integration with 
ogg the baseline video format firefox is supporting. 
http://wiki.xiph.org/index.php/CMML

As mentioned in that blog post she is working on media track description 
language (a early version of that format that we developed is called 
ROE) http://wiki.xiph.org/index.php/ROE
ROE is currently used in metavid to describe multiple temporal text 
tracks, multiple media bitrates, multiple codecs and could be used to 
describe multiple audio tracks. This xml is negotiated by mv_embed to 
give clients the a set of media that fits with their accessibility 
requirements currently based on software platform & supported bit rate 
but could include multiple-language text/audio tracks, audio ~visual 
description~ tracks for the blind,  etc.

==Real World Example ==

MetavidWiki maps these technologies to the mediaWiki.  Multiple audio 
and video mediaWiki resources that differentiate on bitrate, codec or 
language can be mapped to single "Stream" resource which is described 
with ROE. Stream resources are also the basis for temporal wiki text 
data (such as transcripts or subtitles) that could differentiate on 
language or annotative qualities.

When viewing any stream in metavid say the first 10 min of Kucinich 
impeachment proceedings: http://tinyurl.com/52veab
you notice in the little rss-like caterpillar in the upper right links 
to the roe xml representation of that stream:
http://tinyurl.com/4z5mdb
When you embed that clip in a blog for example: 
http://tinyurl.com/4r5xlg all the metadata remains accessible so you can 
access the transcripts by clicking on the little "CC" in the lower right 
of the player. (right now you can only select the English transcript or 
annotative track)...When you click "download" it exposes all the roe 
tracks as downloadable. Notice all the data is temporal and only those 
10min of metadata and media streams of the full multi hour legal smack 
down of GW is requested... 

== Call to Action ==

This is kind of a limited example since its only English... but 
hopefully you get the idea... If someone has an ideal video media set 
with multiple audio tracks & people that could work on multiple 
transcripts I am open to doing more multi-language friendly development 
/ experimentation / demo ;)

peace,
michael


Gregory Maxwell wrote:
> Some folks around here may have input to provide:
>
> http://blog.gingertech.net/2008/09/23/video-accessibility-for-firefox/
>
> _______________________________________________
> foundation-l mailing list
> foundation-l at lists.wikimedia.org
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>