Re: Commercial web site/ILS mash-ups, plus a few observations

From: Conal Tuohy <Conal.Tuohy_at_nyob>
Date: Mon, 19 Jun 2006 14:05:54 +1200
To: NGC4LIB_at_listserv.nd.edu
Tim Spalding:

> So, I'm talking to the tech guy at a library, and I want to get a
> complete list of all their ISBNs. And he tells me that their current
> OPAC doesn't really GIVE them access to that sort of data. With their
> "old" OPAC they could issue a standard SQL command. But now the system
> is locked up, and getting that data requires "buying an XML server."
> To get what I wanted--a list of all their ISBNs--I had to essentially
> *spider* their site.
>
> How common is this scenario?

Pretty common, I would guess :-)

> I'm guessing any discussion of "next generation" OPACs must include an
> open database engine. Am I crazy?

Depends what you mean by "open database engine" perhaps. Though I don't
think it's necessary for an NG catalogue to expose its internal database
_as a database_, nor do I think that'd be a good thing.

The catalogue should expose its data through _bibliographic_ interfaces
and data formats. If a query language is used, it should be CQL rather
than SQL. OAI-PMH and SRW are IMHO the appropriate protocols for
exchange of metadata.

Con
Received on Sun Jun 18 2006 - 22:09:55 EDT