Among much else, Stu Weibel writes:
>
> If we could design a perfect, consistent world, we'd all like there to be
> exactly one, useful-for-everyone, cheap, and optimally efficient standard.
This is indeed possible. The high-end, high-resolution standard being
MARC, with hundreds of data elements specified in minute detail,
fully and thoroughly documented, in use in thousands of applications
with zillions of data, and being known and understood and used the
world over. Moreover, it can be embedded in XML as was already shown
by LC and others. What's needed to make it cheap and efficient is
some interfacing software with APIs in Perl, Java, Javascript, C++
and maybe others, containing functions to convert a MARC XML record
into whatever you want, and concersely, to create one from the stuff
you have. Then, you could also tell a source, hey, gimme the DC
record of xyz, and it would extract it from the XML MARC record.
I mean, it has by now become perfectly clear how much time it takes
and what grueling work it is to arrive at useful specifications for
a standard. But you can hardly name an element that is not already
there in MARC, and documented.
I'm not suggesting, mind you, that everybody learn and use MARC as it
is! With proper interfaces, that's what I'm getting at, it might take
over the role of universal carrier for search interoperability as
well as exchange. The interfacing could make sure that nobody would
ever have to look at naked MARC records or even edit any by hand.
Just as you ought not to have to write raw XML or RDF by hand.
I'm also *not* saying MARC is wonderful and the next best thing to
sliced bread. But what else *is* there to rival it, for the task at
hand?
Even then, however, one would still be left with the incompatibilities
of rules used for the data content. This is likely even the bigger
problem, and for both search and exchange. For this, I have no
solution to offer, but without one there will not be too much
efficiency nor even a lot of usefulness beyond plain keyword
searching. For which we don't need 15 elements, let alone MARC.
Call me backward-minded or hopelessly traditional or whatever, but
not before you prove me wrong or come up with something better.
The MARC community will have to face a major migration, sure, and not
in the very distant future, and for more than one reason. Why not join
forces with them and speed up the whole process? Everybody could win.
B.E.
Bernhard Eversberg
Universitaetsbibliothek, Postf. 3329,
D-38023 Braunschweig, Germany
Tel. +49 531 391-5026 , -5011 , FAX -5836
e-mail [log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|