selecting a search engine for a library web site

[Jennifer_Reiswig at UCSDLIBRARY.ucsd.edu]

     The UCSD science libraries use Verity, which was mentioned by someone 
     from the Verity company earlier in this thread.  To be frank, it took 
     a while to get it running, but now it seems to be stable and working 
     well.  I'm confident that it indexes our pages correctly, and the 
     rankings it produces seem sensible.  I'm not thrilled about the look 
     of the results page, but I don't think we purchased the 
     latest/greatest version.  
     
     Our campus-wide Web site started out with Verity, but found that to 
     get it to run well, we would need to invest a whole bunch more money 
     to upgrade the version we had.  
     
     SO, the campus site now uses Alta Vista as its search engine - they 
     have programmed search box that automatically adds in a string to any 
     user's search, something like +host:ucsd.edu   so that it only 
     searches our sites.  The advantage of this is that we don't have to 
     pay Alta Vista big dollars for this, and it searches all the servers 
     at UCSD, not just the library servers.  If you only want a specific 
     server searched, you can specify that in the string you use - eg, 
     host:scilib.ucsd.edu.  
     
     The BIG disadvantage is that we have no real control over when Alta 
     Vista re-indexes us.  We can ask them to re-index us, but they don't 
     have to comply.  We routinely find entries in Alta Vista searches left 
     over from last summer.  When you follow the link, of course you get 
     the current page, but it may not have those "keywords" on it anymore. 
     Worse, if you've done ANY reorganization of your pages, the link from 
     Alta Vista may be broken.
     
     If you have the money to spend on a real internal search engine like 
     Verity or the others mentioned, I would recommend it - although our 
     programmer might have a different opinion.  Having control over when 
     your site is re-indexed is worth the aggravation of having to keep the 
     search engine running.  The Alta Vista approach is a good alternative 
     though, especially if you don't make changes to file names or 
     directories very often.
     
     Jenny Reiswig
     Biomedical Library
     University of California, San Diego
     jreiswig at ucsd.edu


______________________________ Reply Separator _________________________________
Subject: selecting a search engine for a library web site
Author:  blakely1 at jeflin.tju.edu at @UCSD
Date:    6/24/97 7:40 AM


I am leading efforts to select a new engine for our organizations web 
site.  JEFFLINE [URL:jeffline.tju.edu] currently uses WAIS.  I have 
checked the Web4Lib archives and found little on selecting a search 
engine, most of the materials I found were evaluating search engines such 
as LYCOS, InfoSeek, etc.
     
I have a found a few good articles via an online search but would 
appreciate any help (first-hand experience or good web or print resources.
     
Thank you
     
------------------------------------------------------------------------------
 Elizabeth Linck Blakely, MILS		  | blakely1 at jeflin.tju.edu 
 Information Services Librarian		  | 215/503-7676
 Academic Information Services & Research | 1020 Walnut St., Phila. PA 19107 
 Thomas Jefferson University		  | http://oac1.oac.tju.edu/~blakely1/
------------------------------------------------------------------------------
     
     
     
     
>-- Saved internet headers (useful for debugging)
>Received: from library.berkeley.edu by noc.ucsd.edu; id HAA16678 sendmail 8.8.5
>Received: from localhost by library.berkeley.edu; (5.65v3.2/1.1.8.2/29Oct94-120
>Date: Tue, 24 Jun 1997 07:40:50 -0700
>Message-Id: <Pine.A32.3.96.970602152423.18370B-100000 at jeflin.tju.edu> 
>Errors-To: listchek at library.berkeley.edu
>Reply-To: blakely1 at jeflin.tju.edu
>Originator: web4lib at library.berkeley.edu 
>Sender: web4lib at library.berkeley.edu 
>Precedence: bulk
>From: "Elizabeth L. Blakely" <blakely1 at jeflin.tju.edu>
>To: Multiple recipients of list <web4lib at library.berkeley.edu> 
>Subject: selecting a search engine for a library web site 
>X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas 
>X-Comment: Web4Lib Information - http://sunsite.berkeley.edu/Web4Lib/