what problems have you come up against in embedding the right kind of
metadata for this evolving field? (big question but maybe just one problem
each)
It is a big question and I'm finding this big discussion quite hard to keep
up with. Thanks for your reply Paul Koerbin. Thanks Melinda for assembling
so many interesting people and ideas in one place. However, I guess we just
have to work through the postings and see what happens.
I spend quite a lot of time these days shifting between three zones of
thought when it comes to metadata. These are fresh today because I've just
spent two days in a CORDRA workshop
[http://www.lsal.cmu.edu/lsal/expertise/projects/cordra/] in Melbourne with
many familiar faces and a strong sense of deja vuover all the talk of
metadata repositories and registries. Paul mentioned the notion of common
portals ? this is one of the ideas for how federated discovery and access
might be achieved.
The first thought zone arises from observing the kind of people involved in
the development of metadata standards (there are so many of them now that we
don?t really need any more ? standards that is).
I apologise in advance if anyone takes offence at this (its not personal and
its probably unfair). But it appears to me that most of the major metadata
initiatives (Dublin Core included) have arisen from the anxiety of
mid-career librarians looking for a strategic edge to fend off redundancy.
While the National Library of Australia has some inspirational success
stories in the face of major funding shortfalls, there are some tectonic
shifts happening below the facades of the new or recently refurbished
buildings that are putting cracks in the core values associated with
maintaining cultural memory.
Consequently, these projects are often driven (top down) by imagined rather
than actual need. They are often institution centric and completely out of
touch with the emerging demographic who file share and text each other. I've
long expressed a concern within the Dublin Core community that we were
facilitating the MacDonaldisation of knowledge. In Terminator 2 speak ? the
rise of the machines.
The historical connection between the advance of electronic technologies and
the erosion of cultural memory is cause for reflection. The impact of the
train following the telegraph lines in Central Australia in the 1870s is a
poignant example. According to Daisy Bates (1945) the effects were
devastating.
'With the railway began the extermination of the Central native groups. Each
group through whose territory the track was passing saw its waters used up,
the trees and bushes were destroyed for firewood and fence posts, the whole
country turned to strange uses. They thought that the train and its people
would go away, and leave them the things to play with. They were mesmerised
by the trains, the trains became their life, the rhythm of their days.'
The delivery of the benefits imagined or promised by a technology can also
create a profoundly different receptive environment in which good and bad
can change places according to your point of view. This duality is visible
in the changes that networked technologies have made to the value of the
objects themselves.
The second, arises from trying to understand why so many deployments of
metadata have failed to deliver anything really worthwhile. We can all learn
from failure. If you are interested in a detailed analysis of what went
wrong at ACMI have a look at the abstract or the section: is metadata
harmful?
The third, arises from having to re-conceptualise the whole notion of
information, art, knowledge - how we might think in a world where everything
can be turned into XML. The shape of the container inevitably influences the
way content is both produced and consumed. This can create discontinuities
between different data stores and their production environments. There is a
big difference between the containers needed to hold the kind of atomized
data that fits into relational databases and the kind of information that
doesn?t fit neatly into boxes because it tolerates recursive structures and
the re-use of content along with mixing up the container with the contained.
It?s hard to keep these responses to consumable chunks. It might be best to
stick to stories. I?ve got a great story to illustrate cultural values
associated with metadata?