I'm very pleased to see this discussion, revealing as it does some
differences in perspectives in different communities. These will be
lots of such differences in a cross-domain searching environment. I'm
particularly pleased to see it surface before the CIMI meeting in
September.
I hope discussion there will result in clear distinctions for
distinguishing EVENT metadata from the results or deliverables issuing
from the EVENT.
What do I have when I have retrieved the metadata for an EVENT? A set
of relation elements is one possibility... links to resources associated
with the EVENT? At what point does a metadata description of an EVENT
become a resource in its own right... certainly we have agreed that a
metadata record may stand on its own as a first-class resource.
The challenge to those proposing EVENT as a new Type value is to be able
to make clear in a sentence or two how it should be applied. If it
requires a page of prose to explain, I think it will be trouble.
stu, who is now ruminating over whether there should be a METADATA value
for Type....
> -----Original Message-----
> From: Erin Stewart [SMTP:[log in to unmask]]
> Sent: Tuesday, September 08, 1998 11:32 PM
> To: Alex Satrapa; [log in to unmask]
> Subject: Re: "event" - example
>
> The Olympics is a great generative sort of example from the cultural
> heritage community. Another that may help bring home the importance
> of, if
> not elucidate the semantics for, "event" as type is that of live
> theater or
> concert performances--which people may need to discover by date, local
> sponsors, use of understudies, warmup bands, location, attendance by
> teapot-hoisting dignitaries, etc. I believe that the museum community
> also
> is interested in capturing exhibitions as mutable "resources".
>
> Further discussion of this possible "type" is on the agenda for the
> Consortium for the Computer Interchange of Museum Information DC
> Working
> Group Sept. 21-23 in Santa Monica. I hope that this list will
> entertain
> further comment after the 20 participating cutural institutions from 8
> countries have that opportunity to re-visit this idea and create some
> well-thought out examples for consideration by the wider community.
>
> Erin Stewart
>
> At 9/9/98 09:58 AM, Alex Satrapa wrote:
> >Simon Pockley wrote:
> >
> >> On Sep 8, 1:21pm, Jul,Erik wrote:
> >>
> >> > Perhaps someone who actually wants or needs to create metadata
> records
> >> > for events could enlighten the list with an example and an
> explanation.
> >>
> >> If I could add another example: I was involved in a live cam
> conversation
> >> (CuSeeMe), between New York and Australia. ...
> >
> >Umm... howabout this for an example of an event...
> >
> >In the year 2000, hundreds of athletes from around the world will be
> gathering in
> >and around Homebush Olympic Center, Sydney, Australia.
> >
> >This *EVENT* (all-capitals, flashing lights, whathaveyou) will be
> recorded
> in many
> >forms including video tape, audio tracks and photographs (both
> digital and
> the
> >antiquated chemical paper form ;). People will be taking home
> souvenirs
> from this
> >event, including olympic flags, gold, silver and bronze medallions,
> and
> even blades
> >of grass from the stadium(s? stadii?).
> >
> >A video recording of the Olympics will be just that - a video
> recording of
> the
> >Olympics. The video itself isn't the Olympics. The Olympics will be
> (have
> been) a
> >transient event.
> >
> >To me, the recording of "metadata" about an event seems important,
> but I'm
> not sure
> >that an event can be "indicated"
> (recorded/remembered/signified/whatever)
> by the
> >usual DC set. Who is the publisher of the Olympics? Who is the
> author/creator?
> >
> >You could stretch the model and say that the Publisher of the
> Olympics is the
> >SOC/IOC, with the joint authors being every single athlete, umpire,
> referee, timer,
> >water server, grounds keeper... ad nauseum.
> >
> >If we feel that an Event can be squeezed into our document-oriented
> metadata model,
> >then we must have sub-elements dc.date.start, dc.date.finish (which
> could
> otherwise
> >be the dates on which a composer started and finished a particular
> composition).
> >
> >Just my two bits worth...
> >-Alex
> >
> CIMI DC Metadata Testbed Project Manager
> <http://www.cimi.org/member_area/metadata.html>
> [log in to unmask]
> 972.776.7538 (CDT)
|