Karen,
I'm not denying the possibility of this (it is, after all, roughly
what we're doing here); I'm just skeptical of how it would scale to
larger number of libraries.
Some ILSes don't expose holdings through Z39.50 (Unicorn, for
example). What do you in this case? Even OPACs that /do/ expose
their holdings through Z39.50 might require two requests to get
holdings:
1) for the bib record to retrieve the 001 (or whatever local number)
2) for the opac record to parse holdings.
What I'm getting at is that there are so many variations on how this
would have to be implemented to work for different institutions; I'm
not sure that Z39.50 is any sort of 'key to the castle'.
-Ross.
On 9/12/07, Karen Coyle <kcoyle_at_kcoyle.net> wrote:
> Ross, U of Cal did this in their own link resolver years ago. It drives
> patron-initiated ILL. Sheesh, I remember de-bugging the code!
>
> kc
>
> Ross Singer wrote:
> > On 9/11/07, kcoyle_at_kcoyle.net <kcoyle_at_kcoyle.net> wrote:
> >
> >> but I'm sure that I've been around link resolvers that are at least 5 years
> >> old that would take data from an article in an A&I database (even before
> >> the OpenURL), look up the ISSN in the catalog using Z39.50, and parse out
> >> holdings to let you know if, more or less, your library holds that in hard
> >> copy.
> >>
> > Karen, if you can provide /any/ evidence of the reality of this
> > (outside of one-off local projects like David Walker's at CSU-San
> > Marcos), I will be shocked.
> >
> > -Ross.
> >
> >
>
> --
> -----------------------------------------
> Karen Coyle / Digital Library Consultant
> kcoyle@kcoyle.net / http://www.kcoyle.net
> ph: 510-540-7596
> fx: 510-848-3913
> mo: 510-435-8234
> ------------------------------------------
>
>
Received on Wed Sep 12 2007 - 08:40:23 EDT