David Bearman wrote:
>we are interested in in order to get to it/at it.
It seems as though we are discussing what to do with [off-line] resource types
which, as representations - are themselves ideas.
At Cinemedia we are creating an [on-line] management system to handle
[off-line] representations of ideas such as film, video and CD-ROM. Most of the
information we need to know is covered by the DC elements.
Our descriptions of these physical objects are necessarily coarse because if we
were to try to deconstruct them into their parts or even suggested `types' we
would find ourselves overwhelmed by lists (maybe even frame lists when it comes
to film). A CD-ROM, for example, might contain dozens of different digital
resource formats. It is important for us to know something about these formats
only so that we can discover what system requirements are needed to access this
content. MIME types are useful.
This is jumping away from `types' but I keep coming back to a qualified
`Format' as a way of handling this problem. If it was a video then I guess it
could be:
<META NAME="Type" CONTENT="video">
<META NAME="Format" CONTENT="quicktime">
<META NAME="Format.System" CONTENT="details of system requirements (PAL, VHS) -
maybe a URL">
I'm not sure if this is obvious or foolish but it seems logical to me. It shows
that there is a difference, as David Bearman points out, between getting to an
object and getting at it. Dublin Core appears to be focussed on getting to it
(resource discovery) as opposed to resource access. It would certainly be
useful to hear other people's ideas on why this distinction is (or is not)
important.
--------------------------------------------------
Simon Pockley - Cinemedia (http://www.cinemedia.net)
3 Treasury Place Melbourne Australia [voice] 61 3 96511510
([log in to unmask])
|