Yahoo puzzles me. For the longest time it developed a lot of clearly-related consumer properties — movies, TV, real estate, etc. — and it did things like buy Flickr and Del.icio.us.

But then it released Yahoo Pipes, which programmers have used to create any number of cool data mashups. And then Yahoo announced that it’s launching an open source program to advance the research and development of systems software for distributed computing.

Not that this isn’t cool; it is. Not that Yahoo’s consumer properties aren’t good; they are. It just seems like there are two very different forces of growth here.

Okay, so about this new initiative. According to the press release, the new effort is for levering Yahoo’s leadership in Hadoop, which is an open source distributed computing project of the Apache Software Foundation. (Hadoop’s at http://lucene.apache.org/hadoop/ .

Yahoo’s planning to make Hadoop available (in a data center) for the academic community for software resource. This supercomputing cluster is called the M45 and will run Hadoop (of course) and other open source distributed computing software.

This is a bit over my head. If it’s a bit over yours, you can check out the blog post on Yodel Anecdotal, which gives a little more background information, or the new Yahoo Hadoop blog, which at the moment has only one post.