Thursday, May 28

Google Wave: E-mail re-invented.

Google announced Google Wave at their I/O developer conference.

Techrunch has conference coverage including the Wave keynote. Here is a brief excerpt:
Developed by brothers Lars and Jens Rasmussen and Stephanie Hannon out of Google’s Syndney, Australia offices, Wave was born out of the idea that email and instant messaging, as successful as they still are, were both created a very long time ago. We now have a much more robust web full of content and brimming with a desire to share stuff. Or as Lars Rasumussen put it, “Wave is what email would look like if it were invented today.
Pretty interesting screenshots, check them out. Techrunch also have an exclusive video interview with the creators.

Tuesday, May 26

Distributed Clustering using Mahout on Amazon Elastic MapReduce

Steve Green posted that he got distributed k-means clustering in Mahout to work on Amazon's Elastic MapReduce platform. He wrote up a how-to on the wiki, Mahout on Elastic MapReduce.

R for Data Analysis at Google and Facebook

R is an open-source statistics computing platform.

Mike Driscoll has a post covering the a panel discussion from the Bay Area R User Group meeting with panelists including representatives from Google and Facebook. There is a video of the discussion if you want to watch it.

They highlight the Caret package for classification and regression, rpart for partitioning data sets, and rJava for a Java interface.

If you want to get started with R, a stats background is helpful. Michael Lavine offers his text book, Introduction to Statistical Thought for free which provides an introduction to stats using R.