Re: Categorisation, database searches and RSS format results

[Aaron Swartz <aswartz@swartzfam.com>]

David Davies <d.a.davies@bham.ac.uk> wrote:

> My project for tomorrow is to have the aggregator search across
> databases on different servers.

This is what I see as the possibilities for RSS search engines. Similar to
the work being done with the Gnutella search (previously at infrasearch.com,
IIRC), here's how I see it:

1) A search server allows a user to input a search query
2) The server then calls several CGIs (on various servers) and gives them
the search terms
3) The CGIs return an RSS document with links to the results
4) The results are aggregated and displayed in a consistent way, like with a
normal search engine

Of course, caching would be used to lessen server load and speed results
time.

On Snowball, DanBri's machine, I believe there was a program that used
Sherlock plugins to do searches, but with a web interface. It would
definitely be neat to use this, so that the Sherlock plugins would return
RSS.

This right now, would not be difficult to do, and would be a powerful use of
RSS and a very useful application.

Do others have thoughts?

-- 
        Aaron Swartz         |"This information is top security.
<http://swartzfam.com/aaron/>|     When you have read it, destroy yourself."
  <http://www.theinfo.org/>  |             - Marshall McLuhan