I really think it would be short-sighted to create a system and not pay
attention to the fact that centralized housing is what got us into this
mess in the first place, with both LC and OCLC.
I maintain that any new system for this sort of information must be
built on a P2P system with all the checks that we can for data integrity
(think: BitTorrent). Many Copies Keeps Things Safe, but more importantly
many copies means you can't stop the signal.
No central authority is the future. The information can live everywhere.
Jason
Jason Griffey
Assistant Professor
Head of Library Information Technology
University of Tennessee at Chattanooga
615 McCallie Avenue, Chattanooga, TN 37403
jason-griffey_at_utc.edu :: (423) 425-5449
This correspondence should be considered a public record and subject to
public inspection pursuant to the Tennessee Public Records Act
-----Original Message-----
From: Next generation catalogs for libraries
[mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Joe Shubitowski
Sent: Thursday, May 24, 2007 11:45 AM
To: NGC4LIB_at_listserv.nd.edu
Subject: Re: [NGC4LIB] Authority maintenance (was Subject costs)
My institution develops and maintains three large "authority"
lexicon/thesauri: AAT, ULAN, and TGN which seem to be highly valued by
our communities and receive a ton of submissions for new terms, and
edits to existing terms. It does, though, come down to someone (and
someone can be more than a single entity) somewhere to build and house
the authority. We have a custom built system that we use to build, edit,
maintain, and output these lexicons/vocabulary tools. There aren't
really any off-the-shelf tools that model the necessary workflow and can
deal with poly-hierarchical needs for terms (though WebChoir has some
very powerful tools out now).
Even if everyone is willing to participate in the submission/maintenance
effort for our home-grown LCSH takeover product, it is still going to
require a maintenance agency that houses the database(s), provides tools
and access.
Joe Shubitowski
Joseph M. Shubitowski
Head, Information Systems
Getty Research Institute
1200 Getty Center Drive, Suite 1100
Los Angeles CA 90049-1688
Voice: 310-440-6394
Fax: 310-440-7780
jshubitowski_at_getty.edu
>>> Jonathan Rochkind <rochkind_at_JHU.EDU> 5/24/2007 8:14 AM >>>
I think this is a great idea for a project, but I do not think that a
wiki is appropriate to demonstrate this. You need a real database
environment. You can't do much with controlled metadata on a wiki.
"I'm ignorant enough that the technical question of whether said data in
a wiki setting could be constructed in such a way to allow export of
MARC in a format that can be readily imported into a library ILS is
possible. "
I dont' think so. A wiki is not the appropriate technology for a store
of structured data.
This project also reminds me of William Denton's ideas with "open FRBR".
http://www.openfrbr.org/
As far as setting up a project like this, and the technical
infrastructure for it, there's only so much we can fit in in our 'free
time', I'm afraid. This is not a very difficult project really, as a
pilot, but I wish there were someone applying for grants for it and
such. I wish we had more research in this area going on from library
schools, which are the logical place to house such research, but library
schools seem entirely uninterested in real world library metadata issues
these days.
Jonathan
MULLEN Allen wrote:
> Ed Speer raises some very pertinent questions regarding maintenance of
> authority data. In addition to discussing these, I'd suggest some
> action on the part of the cataloging community.
>
> There is an opportunity now to demonstrate that a collective social
> network of librarians can develop and maintain an authority data set
for
> the series work that LC is no longer doing and that PCC libraries have
> not addressed beyond series they encounter if and when these libraries
> want to trace them (for those PCC libraries that choose to trace
series
> - some, like LC, have decided not to) or that libraries feed to them
(if
> the PCC library staff are open to it and have the resources to do this
> work on behalf of other libraries). Clearly, there is a perceived
need
> among catalogers and, in public libraries at least, among reference
> staff.
>
> I'd like to suggest that a wiki might be constructed in such a way
that
> library staff (and others if they are interested) can enter and edit
> data fields for series authority data. The idea is that anyone can
> contribute a new heading but that the collective wisdom and knowledge
of
> those who know how to construct and document AACR2 series name
headings
> and tracings would maintain the integrity of the data. There is a
need
> - I construct new series headings several times a week when they are
not
> on an OCLC record at all or are in a 490 0 field.
>
> This (or some variant that is technically possible) could be a model,
> albeit with all of the attendant committees and task forces and
policies
> and papers that such weighty matters require in the cataloging world
(as
> well as an enhanced "control" factor that Ed Speer aptly points out is
> necessary), for maintenance of LCSH if and when LC ever lessens or
> relinquishes its commitment to maintenance of the vocabulary.
>
> However, I'm ignorant enough that the technical question of whether
said
> data in a wiki setting could be constructed in such a way to allow
> export of MARC in a format that can be readily imported into a library
> ILS is possible. I'm guessing it is but those of you who are export.
> Certainly, the control fields might be a challenge.
>
> If it is possible, let's do it whether it is sanctioned by NACO,
ALCTS,
> OCLC or anyone else. If it is useful, it flies - if not, it crashes
and
> some other solution to the lack of reliable ongoing series authority
> maintenance for collective use is developed (or, as the case is now,
is
> not). Is protestation and a return to individual library
> decision-making all that catalogers are capable of when faced with
> changes that lessen what gets handed to them on a tarnished silver
> platter? I don't think so - don't mourn, organize to take
> responsibility for what is deemed important.
>
> Allen Mullen
> who is willing to devote some time off-work to helping develop and
> maintain such a system
>
>
>> The problem for an LCSH sans LC is that controlled
>> vocabularies have to be *controlled* somehow. Theoretical
>> rigor with regards to broader/narrower/related headings would
>> be nice as well, but at
>> *minimum* there needs to be a set of agreed upon terms.
>> Otherwise the whole notion of collocation falls apart.
>>
>> Do we invest The Power in a new authority (OCLC?,
>> LibraryThing?) or do we attempt to decentralize? What would a
>> radically decentralized Controlled Vocabulary look like in
>> theoretical terms? (no, tag-clouds don't count) Are there any
>> current examples we could look at for ideas?
>>
>>
>>
>> Ed Sperr
>> Digital Services Consultant
>> NELINET, Inc.
>> 153 Cordaville Rd. Suite 200 Southborough, MA
>> (508) 597-1931 | (800) 635-4638 x1931
>>
>>
>
>
--
Jonathan Rochkind
Sr. Programmer/Analyst
The Sheridan Libraries
Johns Hopkins University
410.516.8886
rochkind (at) jhu.edu
Received on Thu May 24 2007 - 09:49:35 EDT