Re: Why not use DC?

On 18/6/09 11:16, Pierre-Antoine Champin wrote:
> Dan Brickley a écrit :
>> On 18/6/09 02:48, Renato Iannella wrote:
>>> In Section 4.1.2 of<http://www.w3.org/TR/2009/WD-mediaont-10-20090618>
>>> abut 11 of the Core Properties are the same as Dublin Core - why not use
>>> that namespace?
>> One argument might be reduced scope, however in
>>
>> "[Definition: Media Resource]
>>
>>      Any Resource (as defined by [RFC 3986]) related to a media content.
>> Note that [RFC 3986] points out that a resource may be retrievable or
>> not. Hence, this term encompasses the abstract notion of a movie (e.g.
>> Notting Hill) as well as the binary encoding of this movie (e.g. the
>> MPEG-4 encoding of Notting Hill on my DVD), or any intermediate levels
>> of abstraction (e.g. the director's cut or the plane version of Notting
>> Hill). Although some ontologies (FRBR, BBC) define concepts for
>> different such levels of abstraction, our ontology does not commit to
>> any classification of media resources."
>>
>> ...it's clear that a "Media Resource" can be anything, since "related
>> to" is completely unconstrained.
>
> ehr... that was not really the intended meaning of "related to" :-/
> (though now that you're pointing it out, it can indeed be interpreted
> that way).
>
> According to your interpretation, the director of a movie would be
> covered by the definition, which was indeed not intended (or did I miss
> something? ;-)

Yes, I was going on what was written, rather than what I thought you 
might've really meant!

Other projects have danced around this problem. Early Dublin Core talked 
about "document like objects" with a strong emphasis towards online 
documents, but then you get the museums and archival community involved, 
and suddenly a physical painting, or a slide from a photos of that 
painting, or a roll of old movie film, all seem also in scope. And it 
turns out (eg. as concluded by http://dublincore.org/workshops/dc3/) 
that documents like objects, online and offline images, can usefully be 
described with many common properties.

It seems you want to limit things to content / works and content-bearing 
objects, rather than admit the entire universe here?

BTW re "Any Resource (as defined by [RFC 3986])", ... I wouldn't bother 
citing http://www.faqs.org/rfcs/rfc3986.html for a definition of 
resource. All it says is

      "This specification does not limit the scope of what might be a
       resource; rather, the term "resource" is used in a general sense
       for whatever might be identified by a URI."

I suggest instead "thing", since that's all RFC3986 "resource" means 
these days.


> Would "Any Resource (...) representing a media content" be clearer (this
> is a question to the WG as well) ? It may seem a little too "concrete"
> at first sight (e.g. implying "machine representation"), but I think the
> following paragraph makes it clear it can be more abstract.

Taking my "thing" suggestion, that gives ... "Any thing representing 
media content". This seems a reasonable statement of intent. Perhaps the 
best way to make this concrete (amongst ourselves, if not in the spec) 
is by example. Which of the following are intended to be Media Resources:

1. a library MARC record describing a DVD
2a. a HTML page describing that same DVD (some specific DVD)
2b. an Amazon HTML page describing the general class of that DVD (ie. 
that you can buy)
3. An MPEG movie embedding metadata describing the film (from the DVD)
4. A videotape of the film
5. A set of several thousand identical 35mm films with that same content
6. A rusty metal tin on a specific shelf of a national archive room, 
containing one of those films
7. The abstract notion of Shakespeare's Hamlet
8. The abstract notion of the W3C homepage
9. A printed, written transcript of the soundtrack to a film
10. Information about the creator of (9.).

This is just a rough list, from top of my head. Main points: do you 
include physical artifacts directly? does there need to be a digital 
representation involved somewhere?

Thinking again about

"Any thing/resource representing media content", isn't that almost 
circular or somehow redundant, if "media" is taken to mean the carrier 
or means by which content/information is represented or transmitted?

Can you give some more borderline counter-examples (apart from the 
Director) of situations where something is *not* a media object?

Sorry for the pedantry! This is really nice and important work, and I 
especially appreciate the effort you've all put into comparing existing 
efforts...

cheers,

Dan

Received on Thursday, 18 June 2009 09:45:32 UTC