At 11:06 AM -0400 9/8/99, ChaTo (Carlos Castillo) wrote:
> - Faster searches (by including the search engine in a shared object
> library and linking it into Apache)
This is probably not going to be a huge performance win. I talked to
the Apache developers a while back.
> - Better interface (for searching and for BOLDing the keywords in
> the pages that are served as the result for a search)
This is an interesting idea and mentioned often. I hadn't thought
about this particular route, which may be easier than others.
> - Better administrator interface
I don't know that the interface necessarily follows. However, it's a
little easier to hide results forbidden by the server config.
> - Better user feedback (to rate the documents by the number of hits)
I don't know if this would be a great relevance rank. I know some
people who'd hit a page repeatedly to increase the ranking...
> - Larger memory requirements (BUT if the search engine is a shared
> object, the extra memory is shared by all the child servers)
I don't know how to evaluate the tradeoff here. I'd guess it to be a
loss since the hooks for the Apache module would chew up memory. You
might be able to share the engine itself between child threads, but
it still always in memory.
One advantage you *didn't* mention is that a module could more easily
cache common searches.
>PS: please reply cc: [EMAIL PROTECTED] (i'm not suscribed to swish-e
> mailing list).
I think you mean the ht://Dig mailing list. ;-)
If you'd like to discuss this further, feel free to contact me
personally, or we can take it up on the [EMAIL PROTECTED] list.
-Geoff Hutchison
Williams Students Online
http://wso.williams.edu/
------------------------------------
To unsubscribe from the htdig mailing list, send a message to
[EMAIL PROTECTED] containing the single word unsubscribe in
the SUBJECT of the message.
