I've cooked up a little filter for InfoWorld.com's Ultraseek search results. It has always seemed to me that stuff like this should be much easier to find:
So, now they are.
All I'm doing is scraping Ultraseek's results, running a set of classifiers on the URLs and doctitles, and returning results in the same relevance order you'd otherwise have gotten them.
I'll continue tuning the classifiers as I explore the content using this tool; they're very quick and dirty at the moment. And at some point I should transition it to a real webserver. But it's not bad for one modest Python script.
Former URL: http://weblog.infoworld.com/udell/2005/03/15.html#a1194