OPACs on the Web

[hans.schermer _at__ PICA.NL]

[This message is in English. I'm sorry. I can read German, but writing is a bit 
to difficult for me.]

T"ons B"uker implies that Pica OPACs can not be used to drive HTTP servers 
because they do not run on UNIX and they can't handle commandline searches.

I like to point out first that Pica OPACs will run on UNIX workstations. 
Preliminary versions will go into production before the end of the year in the 
Netherlands.

Secondly, all Pica applications are either Client/Server applications or are 
designed with Client/Server architecture in mind. It is perfectly possible to 
write an HTTP server which interfaces with the existing command driven version 
of the Pica OPAC on the basis of the well defined Pica-3 protocol. The platform 
is not relevant as long as a C/S connection can be made. IP or even a serial 
line can be used to link to the Pica OPAC.

However, it is much more useful to write an HTTP server which communicates with 
Z39.50 services. This opens up much more sources of information to the user. 
Through a server like that, a library could combine OPACs from various 
libraries, with locally mounted reference databases and OCLC's Firstsearch 
services, all with the same simple user interface. The next step off course is a
WWW client with Z39.50 support. This will give total control to the end user. In
order to support and develop this technology Pica is developing Z39.50 access 
from and to both local and central systems. I hope the allegro system will allow
for Z39.50 access also.


Hans Schermer.
hans.schermer _at__ pica.nl
+31-71-257168
(I work at Pica in Leiden, but my opinions are my own and not necessarily 
Pica's)