FeedMiner: A Work In Progress I Hope

I was excited to see a new feed finder, hoping they were perhaps onto something the others have missed. At this point it seems FeedMiner has missed it too.

There's no ubiquitous "BETA" tag anywhere so either its creators are as sick of seeing them as we are or this is meant to be a finished product.

I tried two searches one specific and one general. I chose "fantasy baseball" ('tis the season) and "technology." Results for "fantasy baseball" were:

1. Fantasy Baseball : CBS SportsLine.com Fantasy Sports: Good!
2. Velocity Sports Blog: Sports, yes, but nothing about fantasy or baseball.
3. Fantasy sports podcast: It would be nice to keep podcasts and news feeds separate here, but at least this is somewhat relevant, if not specific to baseball.
4. McSweeney's: What? Not even close.
5. Baltimoresun.com: On Fantasy Sports: Not bad. Specific to the Baltimore Orioles mostly, but the columnist also discusses the rest of the league. Not Top Five-worthy though.
6. Will Carroll Presents: Will Carroll is a baseball writer, yes. However this feed appears to have expired two years ago.
7. SI.com - Fantasy sports: Good!
8. Amazon: baseball: This is an Amazon.com feed for books about baseball.
9. Wisch List: This is a non-sports Chicago Tribune Columnist.
10. Copyfight: A legal blog.

So I'd give those results a 3.5 out of 10 for accuracy. I would do a lot better if I did a regular Google search for "Fantasy baseball" and then checked if any of the sites had RSS feeds. Pretty poor.

For the general search "technology" the results were a bit better but I was alarmed to see the #1 result was a spam site. The rest of the results were largely the technology feeds for major news sites. Not bad, but hardly interesting.

Needless to say, I'm unimpressed with Feedminer's results. It also offers no bells an whistles like exporting results as OPML or saving good results to a personalized page. I have nothing against the stripped-down Google approach to search, but the results better be damn good.

Pipe Down: Fun with Yahoo! Pipes

Well, I've finished my first Yahoo! Pipe. I decided to create a feed that pulls from several "life hacking" blogs and limits results to posts containing lists of tips (which I often find myself drawn to). I rounded up feeds from a list of popular life hacking blogs like Lifehacker, 37 Signals, and WebWorker Daily and told Yahoo! Pipes to Fetch them. Then I connected that to the Content Analysis operator, and then connected that to a filter. I had to set the filter up to permit only blog post titles that contain keywords common in "Tips" type posts. I chose: "tips, top, some, favorite, best, steps, and ways." This was based on reading through several of the blogs and noting what words commonly appeared in the titles of "tips" type posts.

Lastly, I told it to sort by Publication Date and filter out duplicates. Here's the result:

http://pipes.yahoo.com/pipes/xMKQRnW92xGrLKYelvXiAA/run?_render=rss

It worked pretty well. A few non-tip posts made it in, but for the most part it did what I'd hoped. I remain very excited about Yahoo! Pipes and will mess with it as much as time allows. It's still beyond non-techies, but I can see this becoming a powerful tool for Webmasters who want to display targeted content and power news junkies who want to create truly unique "newspapers." Yahoo! Pipes plus a newsfeed aggregator like Netvibes or Protopage equals the world's greatest online newspaper.