The latest version of Dexter walks faster and more robustly.

Pretty impressive! However, regarding the practical prospects of humanoid / antropomorphic robots I remain pretty sceptical. Why? Because wheels (or sometimes threads) are so much simpler and cheaper! It seems to me like a classic case of "Worse is better".

The usual arguments for legged robots aren't very convincing, IMO:

- "legged robots are better for difficult/uneven terrain": look at some videos showing the iRobot PackBot (which runs on threads): Although it's very small, it can climb stairs.

- "humanoid robots have an advantage because people can interact more naturally with them": maybe, but people also interact with Roombas, dogs, cars etc. - none of which are humanoid. Regarding the Roomba, I have read an article that many people name their Roomba and even pity them if it gets stuck. So the question will be: is the advantage of "better interaction" worth the considerably higher price for a humanoid robot body?

- Isaac Asimov said that humanoid robots will have an advantage because the whole world is already build for humans. Again, maybe - but when the car was introduced, we rebuilt the whole infrastructure (i.e. streets) to meet our new machines' needs.

Perhaps I'm just not getting the point of humanoid robots - apart from beeing technically impressive and cool to watch.

As Trevor was kicking his legs and pushing on his knees, I was waiting for Dexter to turn around and slap him silly. Guess I've been reading too much SciFi.

Just wondering: Do the people in robotics realize that the real money is in getting machines to pick cabbages and the like? These sorts of inventions (like any others) really need to replace labor to be economic.

"Do note that the "total" time spent (e.g., time spent in Ruby, instead of waiting on IO) is much lower for Sphinx than for Solr. For both of them, the socket overhead is a constant penalty, and won't increase."

Hey, why not write a pure Ruby indexing and search engine?

Oh right, because Ruby is too slow.