Print

Print


> -----Original Message-----
> From: Z39.50 Next-Generation Initiative [mailto:[log in to unmask]]On Behalf Of
> Eliot Christian
> Sent: Thursday, July 22, 2004 7:25 AM
> To: [log in to unmask]
> Subject: Re: Demo business case for search interoperability
>
>
> At 08:42 AM 7/22/2004, you wrote:
> > > Date: Thu, 22 Jul 2004 08:13:48 -0400
> > > From: Eliot Christian <[log in to unmask]>
> > >
> > > As a next step following documentation of the requirements for search
> > > interoperability ( http://www.search.gov/interop/requirements.html ),
> > > I need to construct a simple demo showing what benefits one gets
> > > from standardized search services. The task is to do a side-by-side
> > > compare of information searching: "with a standard" versus "without
> > > a standard".
> > >
> > > My immediate thought is to use a metasearch example, simply because
> > > metasearching by definition has to interface with multiple search
> > > service instances.
> >
> >How are you going to do the "without a standard" version??
>
> Run the equivalent searches sequentially (e.g., Google search Web
> pages, LOC search of catalog records, ...) using their separate
> user interfaces. The final step in this version is then to do an
> "integrate-by-hand" of the results sets.
>
> Of course, if the end goal is to achieve something like "result
> clusters within facet", then this final step is Really Difficult
> (ergo, a key selling point of the standard search service).
>
This is confusing two aspects. You are actually contemplating a "federated
search" ("metasearch") vs. a "sequential search". The outline has nothing to
do with use of standards or otherwise, it is purely the search workflow. Of
course this is actually where the biggest productivity gains are to be made,
but it does not depend on 'use of standards'.

Your "without a standard" examples would be difficult to do *with* a
standard as Google does not have a "standards based" interface (assuming for
the sake of this argument that "standards based" means using a search
standard like Z39.50 - http is not really a comparable standard!).

The Library of Texas system is a metasearch system using Z39.50 as its
search standard. A comparison to this would be to run the same searches
using a metasearch engine which is capable of running searches using
different "protocols" and connecting to different interfaces. Two examples
of these engines are Muse (from www.museglobal.com - my company) and Webfeat
(from www.webfeat.com). Using these you could compare federated searching
and sequential searching, and standards based (z39.50) vs. non,
independently.

We have some demos you could have access to - contact me if interested.  -
Peter Noerr ([log in to unmask])


> > > Please do let me know of any canned demos you may have on the shelf,
> > > or any ideas on how best to pitch this. (Of course, hard
> > > cost/benefit data would be nice if you have that, too!)
> >
> >I imagine that the Library of Texas people would be happy for you to
> >use their system for a demo.  The main people to talk to would be
> >Kevin Marsh <[log in to unmask]> and Bill Moen <[log in to unmask]>;
> >I've copied this message to them.
> >
> >Hope it works out!
>
> Thanks!
>
> Eliot
>