[WEB4LIB] "search our site" stats

[Avi Rappoport]

It makes a lot of sense to me, most end-users have a hard time 
articulating the kind of search they want to do.  Amazon has gotten a 
lot of praise for having a simple search field that distributing 
queries to multiple backend search engines and databases.

You could do this with Z39.50, Muse Global, or Harvest-style systems, 
metasearch middleware systems, or just write a script that sends the 
queries and displays the results.

I have some slides from a talk on P2P, metasearch and distributed search at
   <http://www.searchtools.com/info/slides/searchengines2002/>

Someday, I'll get the article finished as well.

Best of luck,

Avi

At 10:49 AM -0800 3/25/02, Blake Carver wrote:
>A while back there was a bit of a discussion on the pros and cons of a
>"search our site" search engine. At CIL in DC I noticed a couple of people
>say that people don't seem to make the distinction between a "site search",
>an "opac search" or a "database search", it's all just searching the
>libraries web site to our users.
>Has anyone else been keeping any kind of logs?

-- 
Complete Guide to Search Engines for Web Sites and Intranets
    <http://www.searchtools.com>