Andrew Marlow wrote:
> This is fantastic. I want it!
Thanks. I'll note the support when I propose working more on
it next year... :-)
> Hmm. Whilst your link shows some great categorization, stuff local to
> Penn might be a problem when it comes to international systems (I am in
> the UK).
Let me clarify: the classification system is *not* local to Penn; it's
the Library of Congress subject headings (LCSH) ontology. (Which is a bit
US-centric, but not so much that it would seem very odd in the UK,
I'd think.) The "local to Penn" bit simply refers to the *data* I
used, which was a copy of the LCSH ontology made and adpated by
Penn some years back, so it's somewhat out of date and doesn't
completely match current LCSH.
The LCSH subject headings structure can be searched from
http://authorities.loc.gov/
and I was going to also direct folks towards a wonderful downloadable
SKOS version at
http://lcsh.info/
but the LC staffer unfortunately has just had to take
it down a few days ago at his employer's request.
Annoyingly, I had not yet saved a local copy.
(Did anyone else? Please write if you have; I'd love to get a copy.)
At the moment, the best available free alternative appears to be a copy
of the subject authority records (in MARC XML, not SKOS) scraped off the
authorities site a couple of years ago, which you can find at
http://www.ibiblio.org/fred2.0/authorities/
Note that the LC might claim copyright in these records outside the US;
I'm not sure about this. (Inside the US, they can't claim copyright
to their own work, since they're a federal government agency.)
***
By the way, if anyone's still interested in a using Dewey decimal-like
system, but finds the 1876 version cited earlier too creaky,
there's a later revision from 1922 (the 11th edition) now digitized
at Google. I've recently added a listing for it at
http://onlinebooks.library.upenn.edu/webbin/book//lookupid?key=olbp44744
(Non-US users might need to use the archive.org link I provide
there instead of the Google link, but it's public domain just about
everywhere, since Dewey died more than 70 years ago and the
1922 copyright has also expired in the US.)
It looks like the 12th edition, from 1927, is also public domain due
to non-renewal, but I don't know of an online copy of that one.
(The 13th edition, from 1932, is posthumous, and its copyright was renewed.)
If you were using it in a repository today, you'd need to update it
to include things that Dewey somehow failed to notice in 1922, like
computers, and you'd probably also want to undo the "reformed" spellings
that Dewey was keen on at the time. You may also want to be aware that
some topics in use then have been moved around in the current Dewey
Decimal Classification (DDC, 22nd edition). Subjects like telegraphs and
canal engineering somehow now don't seem to warrant taking up as much
of the numeric space as they used to.
As Ralph LeVan points out, OCLC claims copyright on "the text associated
with the numbers". So you can't just copy the current DDC guide without
getting permission and agreeing to a license. But I don't know of anything
stopping you from looking through local library stacks to come up with
your *own* description for what the library is now putting where. (I'm
not a lawyer, but my understanding is that, at least in the US, reverse
engineering for compatibility or interoperability with an existing system
is generally not considered copyright infringement. And if you're planning
on interleaving papers deposited in your IR with other library resources
in a discovery interface, compatibility is a legitimate concern.)
[Note that if you do re-engineer a schema, you shouldn't call it
"Dewey decimal", because it wouldn't be exactly the same as current or
past DDCs, and because OCLC has the trademark for that term. But it might
be close enough for your purposes, if you're keen on the numeric-code
style of subject browsing, and licensing DDC or UDC won't work for you.]
John
------------------------------------------------------------------------------
_______________________________________________
DSpace-tech mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dspace-tech
