[Ffmpeg-devel] samples database

[Mike Melanson]

Attila Kinali wrote:
> That would be an idea. Though not very scalable, but it has the
> apeal of a simple solution. 
> 
> Anyone out there who would like to do that job?

The idea depends of capabilities of MediaWiki that are of uncertain
existence (like per-page ACLs). I just thought of a possibly better
refinement: maintain an actual textual CSV file and store it under
revision control. Then, the people with permissions could edit the file
in a tool that is significantly more conducive to CSV data, like a
spreadsheet (OpenOffice Calc or MS Excel). Host the CSV file wherever
via HTTP and create simple tools in whatever very high level language to
query it. Update the live web copy whenever a commit is made (and an
integrity test is passed). Compress the file of the server for optimal
transfer.

Another lazy, pragmatic solution.

-- 
	-Mike Melanson