Repost - Search engine for local HTML files

[Michael Collier <mcollier -at- ARLUT -dot- UTEXAS -dot- EDU>]

I'm reposting this due to the small number of responses to the original post
and the large number of requests for further information. I know there are
people out there who have experience with this topic. Please post to the
list.

-------------------------------------

Having converted about 600 Unix text files (in about 11 subdirectories) to
HTML for internal web site posting, now they need to be searched from the
browser. Since I'm not sure how to best proceed I thought I'd see what ideas
the list may have.

I'm still in the process of collecting information on how users will search
the files, but I also want to consider now how to implement the search
function. Possibilities include:

Server side--
Web server search engine module--need to see if this is set up; not sure if
I have access to it.
Third party search engine (binary)--I may not have full access to the web
server, plus there are platform issues (Unix, NT users)
CGI script--again I may not have full access to the server, but if so would
this slow down the web server too much, given the number of files?

Client side--
Use a Java applet (or can this be done effectively with Javascript?). The
site visitors all have Java-enabled browsers. However would I have to spend
a lot of time reindexing the files for this solution? (Although I don't
think the files will be changing too much).

Externally hosted search engine--
Excite, etc. but I don't think I want their ads, etc. all over our pages and
I don't know if I can control navigation between our pages and the search
engine.

What I've already got to work with, if needed: a 60980KB Access database of
all the words in the HTML files, and a 24134KB HTML index file (these were
accomplished using a shareware indexing tool).

I'd appreciate hearing any thoughts on the best way to set up the search
function.

Thanks,


 From ??? -at- ??? Sun Jan 00 00:00:00 0000=