Re: Authority maintenance (was Subject costs)

I believe the first use (vanilla authority files) needs to be primary
albeit with a far more open structure than the LC/NACO/OCLC model.

 However, with the buzz of the hive, series authority records could to
(and potentially, as you suggest) back from librarycatalogs, from
WorldCat records for the individual books, incorporate reviews and
tagging, link to web sites for fans of the series (often on publisher
sites), link to order information, provide readers' advisory (if you
like this series, see...), to tables of contents and citations for
issues of professional monograph series, to Google books, and more that
all of you might imagine.  It could demonstrate the power of social
networking in all kinds of ways.  But the primary level needs to be
there and needs to be reliable, methinks!

Thanks again for your thoughts. Keep brainstorming.  Allen

>-----Original Message-----
>From: Next generation catalogs for libraries
>[mailto:NGC4LIB_at_listserv.nd.edu] On Behalf Of Karen Coyle
>Sent: Friday, May 25, 2007 8:10 AM
>To: NGC4LIB_at_listserv.nd.edu
>Subject: Re: [NGC4LIB] Authority maintenance (was Subject costs)
>
>While we're brainstorming....
>
>We've been talking about an authority file, but not how we'd
>use it. It could be like LC authorities, that we use for
>"copy" -- that is, when we need to add a new entry to our own
>catalogs, we ingest that entry from the community file. This
>is a use that would require minimum support in that the file
>is used once by a library for new entries. The library system
>would then do the indexing an display that interacts with the
>user view of the bibliographic data.
>
>The other use would be if we actually pointed our catalogs at
>this file and somehow used the data dynamically in the user
>view. This would be a much heavier use of the system, and
>would require a great deal more cycles and bandwidth.
>
>I presume we are talking about the first use, a large database
>of authority "copy" that can be updated by librarians but that
>is mainly used to download data into library catalogs. This is
>basically the NACO model, although in that one a small set of
>elite libraries is authorized to add and change records. And
>there's no reason why this "wiki" model couldn't limit input
>to a trusted set of experts. Or at least limit the "record"
>portion to experts, and allow other users to have a sandbox
>for whatever information they wish to add.
>
>I still don't see it requiring a whole OCLC-like structure for
>hardware and staffing. It'll be a reasonably big database, but
>the hits against it will be somewhat low... unless it ends up
>appearing on the first pages in Google searches.
>
>kc
>
>