Daniel Aleksandersen wrote:> Hi list,
>
> On 2007-09-10 Silvia Pfeiffer wrote:
>> 2) we need a XML annotation format for audio - in particular for music
-
>> that is more structured than vorbiscomment (and this probably
>> applies to video, too)
>
> Above, Silvia sums up nicely what I used three days and a tone of emails to
> say. (Though she did not say that this format should replace Vorbis
> comments all together, which is still my view.)
>
Comments have the advantage that they're lightweight and easy
to support. Once you include XML parsing in the equation you
require a much more complex framework to parse and present
that information. I think Vorbiscomments still have
applications, for hardware players and for more lightweight
software players. It is also more simple to add Vorbiscomments
at the encode stage and the full metadata later. Ideally they
will be subordinate to proper metadata.
>
> What have been done so far is to create a wiki page with my original XML
> based metadata format[1]. This format needs much work.
>
> I think the very next thing to be done is to define exact what the format
> should and must not describe. My original proposal were something along the
> lines of: ?[The format] should give a metadata description of any media
> resource (text, video, audio, ...) in an Ogg stream; for use with search-
> and multimedia management software.?
"should" is fine, but 'must not' is too much (I'm being
RFC-
strict here). You're right, work out what we want to describe.
But I still believe leave room for others to add their own
metadata (it's where we'll get strong use cases showing up),
as well as allowing room for further Xiph additions without
needing to start a new spec. This approach also lets you
focus on one segment at a time. "Search and multimedia
management software" puts us into indexing.
>
> I doubt this will be the final definition of what the format does, but it
is
> a start. Please do send any definition suggestions or thoughts to the email
> list so that we all may better understand what will be required by the
> format.
>
For photography maybe look at what EXIF provides. Someone
needs to work out what XMP provides.
http://support.adobe.com/devsup/devsup.nsf/xmpkb.htm?OpenView&Start=1&Count=500&Expand=7#7
> Regarding the poor naming for my proposal:
> I just though up a name for the XML format; and MDMF was the best I could
> come up with at 5am. I think no one will oppose changing the name to my new
> suggestion (right below). Though someone might be better at naming stuff.
>
> M3F, or the Multimedia Metadata Format.
>
> Thoughts?
>
"What's in a name? that which we call a rose
By any other name would smell as sweet;"
It doesn't matter too much. It need not even be the one
eventually used for a magic number in embedding (though
that would make the most and the use of a different
'name' as the mn would need a good reason). I did have
a couple of non-acronym ideas kicking about but naming
is a privilege of the person who actually does the work.
>
> [1] http://wiki.xiph.org/MDMF
>
> (How am I doing with the leadership-thing, Silvia?)
The first rule is to never ask...
--
imalone