I added a search engine to this site tonight. The engine is a program called swish-e; it was interesting to use, because it has the ability to index the contents of a database, which is where most of this site lives. It does a better job than just a straight SQL search, too, because if you enter more than one word for a SQL search, at least the way I used to do search on Geneablogy, it would search for the whole phrase rather than instances of any or all of the words within the entire post.
I've also added tags to the head of the main page that link to my RSS feed and an OPML rendering of my blogroll. I'm not sure what having the blogroll available in that way will do; I'm still trying to get my head around that. Something about showing what neighborhood you're in based on who you link to and who links to you. Since nobody links to me, I must be out in the sticks. But it was a 20 minute exercise to implement, so what the heck. Putting the RSS feed in a link tag does something called RSS Autodiscovery that I'm also not completely clear on, but it makes it more semantically clear that there's a link to an RSS feed for anyone who cares, so I figure it's a good idea. This stuff seems to be moving with dizzying speed....
Posted at 12:56 AM
Note: I’m tired of clearing the spam from my comments, so comments are no longer accepted.
I've got swish-e set up so that it only runs whenever I publish a new story. The code I use to tell the database to go ahead and publish also contains a call to swish-e. That way the index is always up-to-date and I don't have it indexing unnecessarily. No cron involved.
That said, on the anarc.org site, I've never had a problem with the cron job that runs swish-e dying.
Posted by ralph at 12:32 AM, June 5, 2002 [Link]
This site is copyright © 2002-2024, Ralph Brandi.
I used swish-e with the Naaso site. It worked really well so long as your cron job updating the index actually worked. Mine kept breaking for some unknown reason.
Posted by Little Brother at 2:56 PM, June 4, 2002 [Link]