Hello!
Ian Stuart wrote:
> In my view, unless you build the hierarchical SWAP relationships from
> the outset, you will never actually be able to build a proper SWAP tree
Funny how we're back to needing hierarchical data models. They went out
of fashion a while back... Nothing is ever new eh? ;-)
What worries me about the hierarchy is how do we build that hierarchy in
the first place? If people cannot even manage to add a couple of subject
terms - controlled or not - then how can we ever expect them to specify
"this conference paper is part of this work"?
Well, hide them having to do that with a nice interface (snazzy drag and
drop AJAX-based - just imagine it!) is one way. But I'd suggest
repository software development suffers a bit from what you might call
the "developer-bewilderment". Some software developers will say "Google
does discovery without metadata" and think that means metadata,
controlled vocabularies, structures like FRBR (and thus SWAP) are at
best unnecessary and at worst just plain wrong.
Its a difficult area. If software developers see Google and automatic
classification as saviour - which is no surprise as these people (me
included) are excited by the technology - they'll not want to build nice
interfaces to create complex, structured metadata. Add to this the fact
that Open Access evangelists are saying things like "don't bother with
subject classification, no one *searches* using that" (seeing, probably
rightly, that people wont self-archive if they need to include lots of
metadata) and there is a trouble.
Seems to me such views are opposed by the library and information
scientists who, rightly, say we need metadata and there are useful
things that can be done with subject classification, structured
metadata, etc. It is important to note that the things supported by
metadata are not just about discovery *now* but in the future. (Consider
the classic example of "The Great War" and "World War I" - it only
because WWI when WWII happened and before that no one described it as such).
Of course, I'm generalising extreme views here. Point I'm trying to make
is that we need to avoid both extreme if we're to get anywhere and get
there in a sustainable, long-term, way.
Pete Cliff, Research Officer
UKOLN
|