Once again, I’ve been tinkering with the search on this site. We’ve gone through several iterations here, but I think this one is a keeper.
Most recently, I abandoned the SQL-based search system I was using and switched to using Google.
However, much to my surprise, it didn’t work that well. It didn’t return results for terms I knew were in the page I was looking for. I’ve done a little comparison between Google and the new system, and it’s confirmed my suspicions — there are several instances where Google just didn’t pick up on words. Go figure.
The new system is based on Swish-E, which is a phenomenally well-done, open source, indexing engine. It updates once an hour (swallowing 4,500 HTML files in 8 seconds) and it seems to just work, which is refreshing.
Try it out, and let me know if you have any problems or suggestions. I have some pretty cool refinements for the system coming, but I want to get it “in the wild” for a month or two before I start tinkering.
I've been toying with an idea lately, and instead of actually doing it (don't have the time), I'm going to throw it out here for fun. My idea is for an extremely simplistic content management system -- one based on HTML files and a scheduled file system crawl. First, some…
I worked on improving the search on this site today. Search has been through a number of iterations. First, I used the basic Movable Type search. But it was slow and I wanted to do some interesting things with search. So last year, I switched to using a SQL…