On Wed, 18 Aug 1999, Maxwell Gwynn wrote:
>
> I have recently installed "Excite" (www.excite.com) as my homepage (it's
> free). They have a news-clipping service which allows one to
> enter any number of keywords (mine are, among others, "psychology,"
> "hypnosis," "eyewitness," and "Montreal Canadiens" ). I imagine that
> other search engines (Canoe (CANadian Online Explorer), Yahoo, AltaVista,
> etc.) have similar services. I'd appreciate hearing about other TIPSters
> experiences with these.
The _Science_ column "Netwatch" for July 16/99 (vol. 285, p. 295)
recently reported a study of search engines. They reported that no
single engine covered the web very well, but in top place was one
called "Northern Light" (http://www.northernlight.com), which covered
16% of the web. Excite was way down on the list, at 5.6% coverage.
These were for single search engines; unfortunately they didn't rate
search engines such as metacrawler which combine a number of search
engines and in theory, at least, could do better.
Although I usually use metacrawler, I gave Northern Light a try, and
I'm very pleased with the result. It seemed to me it pulled out more
useful sites than when I searched using metacrawler, although I didn't
do a systematic comparison. In particular, it has its own private
collection which it offers to sell. That seems to include a good deal
of the published scientific literature and a variety of other
documents not available on the web, and includes biblliographic
information, so interlibrary loan can be used instead. It also has a
nice feature of dividing searches into categories ("folders"), which
allows you to concentrate on hits which are the most relevent for you.
-Stephen
------------------------------------------------------------------------
Stephen Black, Ph.D. tel: (819) 822-9600 ext 2470
Department of Psychology fax: (819) 822-9661
Bishop's University e-mail: [EMAIL PROTECTED]
Lennoxville, QC
J1M 1Z7
Canada Department web page at http://www.ubishops.ca/ccc/div/soc/psy
------------------------------------------------------------------------
