I’m still playing around with the standing search idea to see if Feedster or Blogdigger or some other tool stands out in terms of the results I get via the RSS feed. The main requirement right now is not getting duplicates, followed by a greater level of relevance. If there’s one thing that I’ve been disappointed by it’s that relatively few of the hits I’ve been getting back are of any use, and try as I might to tweak the terms, it’s not getting much better.
PubSub has come out with a new iteration, however, that might be the better answer. I’ve been tracking two searches for the past week, and even though the number of results has been relatively small, the relevance is way up. I added a couple more today that I am also tracking in Blogdigger to serve up a comparison.
What’s good about the PubSub search is the many ways you can use boolean syntax to narrow your search. I need to do a bit more digging, but on first blush, I don’t think the other two do. And, PubSub gives you the option of just searching a certain percentage of the 3 million or so blogs in their database, based on how popular a site seems to be on any given day. That’s another great way of limiting the results. All in all, more choices, and choices are good, right?
Blogdigger’s query syntax supports boolean operators such as AND, OR, NOT +, -, as well as fuzzy operators, wildcard operators, range and proximity searches. A good overview is available at the Lucene site (Lucene is our query engine). Additionally, you can specify which fields to search in, for example, title:education searches only the post title, subject:education looks in the subject field, author:will would look in the author field. If I can help you construct any other complex queries, let me know, and I’d be happy to help out.
Thanks again for using Blogdigger!