[WEB4LIB] Re: Z39.50

[Walt_Crawford at notes.rlg.org]

Dan asks:

>I also understand that a number of ILS vendors are moving away from
>using Z39.50 as an intermediary between their native OPAC and their
>web OPAC.  Walt, what figures do you have on that?

None, I'm afraid: I don't have any inside information on the library
automation industry (except our corner of it).

My naive sense is that the distinction between "native OPAC" and "web OPAC"
is itself dying--but that may have to do with the libraries I use. If the
primary user interface is a Web-accessible interface (as seems to be
increasingly the case--unless I'm just out of it, as is possible), then
efficiency and special-hooks reasons might well cause a vendor to use
something other than Z39.50 as the communications mechanism.

Which is to say: I think those figures would be fascinating, and would love
to see comments from vendors out there.

What we have seen here is that our Web Eureka represents a growing
percentage of all apparent end-user searching--but that's at the expense of
the telnet interface, not direct Zephyr/Z39.50 use. The latter (as a
percentage) had been declining in 98/99, but seems to have recovered in
00/01--probably because two or three major institutions went to local
systems that did a good job of supporting Z39.50. (Again, I'd love to see
OCLC's figures, if they aren't proprietary. ALL of our end-user searches
are Z39.50 searches; it's just a question of whether the client is one of
ours or a library's own.)

I think there is an anecdotal case that standalone Z39.50 clients are
declining in use, for some of the same reasons that home-grown library
catalogs are declining in use.

None of this is authoritative, to be sure.
-walt crawford-