If you haven't see Isite/Isearch yet, that is a free
text indexing and information retrieval software product that
implements Z39.50 service protocol. http://www.cnidr.org/ is its home.
Source is C++; it is an outgrowth or next-generation (last generation?)
of WAIS-like systems.
Outside of use of Z39.50 by some library
catalogs and goverment databases, I haven't seen much use
in recent years of this information retrieval
system, and user clients for it are rare or no longer
much developed. It is a valid Internet IR protocol, but
HTTP/web stuff has essentially superceded it. In general you
can do same + more things with http client/server than Z39.50.
Z39.50 is a fairly complex protocol, and lacks provisions for
browsing/information discovery (it is mainly a search/retrieval
protocol). Don't let me dissuade you or others from investigating
it; the Isite framework would make a good foundation for
adapting Z39.50 to other uses.
My own feeling is that CORBA may
have more life left in it as a client/server method other than
HTTP for bioinformatics information retrieval and related projects.
CORBA isn't a specific IR protocol like Z39.50, but it is flexible
enough to be adapted to that, plus allow more specific data and
method interactions for bioinformatics projects. It also includes
service discovery methods (forgot the details of this...).
In article <003f01c020b5$43120e00$7b02a8c0 at nic.in>,
Malay <curiouser at ccmb.ap.nic.in> wrote:
>>Does anyone have any idea whether Z39.50 has been implemented in any
>software in Bioinformatics (exclude commercial reference managers). I'd be
>glad to have access to any documentation, library, partial or complete code
>listing in any language.
-- gilbertd at bio.indiana.edu