[WEB4LIB] Re: How are people using your "Site Search"?

[Tim Smith]

I agree with Blake Carver's and David Merchant's observations about patron
(mis)use of site search engines. Like David, we've been using Atomz.com's
search engine (though not for free; we have way too many pages for that),
for a couple of years now and, like him again, I have tried to explain what
the search engine does and does not search:
http://www.library.ohiou.edu/electres/search/asearch.htm. I think the
"disclaimer" has helped some--but it's never going to overcome the problem
of people (like me, in other contexts!) who don't read instructions.

I also thought it might be interesting and instructive to share the list of
top searches from our log at Atomz. Here it is:

Feb. 10-16

8    lexus nexus
7    lexis nexis
6    psycinfo
5    electronic reserve
5    symbolism
4    free radicals
4    movies
4    sport management
4    standard and poor s bond guide
3    a measure of workplace
3    acedemic advancement center
3    college basketball
3    computer lab
3    eric
3    flic
3    friends of the library
3    genealogy
3    harry potter
3    lexis nexus
3    meigs county
 


Many of the searches are reasonable for a website search, while others
clearly are not. One of the very nice things about the Atomz search engine
is that it allows the administrator to insert "synonyms" in the
configuration of the system. Which is to say, I've made lexus and nexus
"synonyms" for lexis and nexis, which allows all those misspelled searches
to come up with good results. Likewise, psycinfo, psychinfo, psychlit, and a
few other sets of similar words. (In fact, I see from the list above that I
should probably add "acedemic" to the set of synonyms.)

Tim Smith

*     *     *     *     *     *     *     *     *     *     *     *
    Tim Smith                    Phone: (740) 593-2634
    Reference Dept.              E-Mail: tsmith1 at ohiou.edu
    Alden Library, Ohio Univ.    Fax: (740) 593-0138
    Athens, OH 45701

    "Technology has replaced reflection" -- Utah Phillips
*     *     *     *     *     *     *     *     *     *     *     *