Dynamic langages like Python or Ruby are considerably easier than static languages like Java and C++. John is asking us why:

How do you account for the huge increases in productivity that dynamic language advocates say they experience.

My answer is duck typing. It is powerful because formal and tight definitions are hard and less useful than we might expect.

The difference is fundamental:

Further reading:

• My answer to John on Stack Overflow.
• My post Duck Typing, Artificial Intelligence and Philosophy.

Julian Hyde just announced that Oracle will support MDX: they were the last vendor to resist this Microsoft technology. MDX is to multidimensional databases what SQL is to relational database. 

I am disappointed that MDX is the best we could come up with. I don’t find MDX enjoyable.

Further reading: What are the computer langage people waiting for?

One of my research interest is tag-cloud drawing. The domain is a natural extension to the classical Graph Drawing problems. However, it is much more open-ended. I keep seeing new types of tag clouds each year!

Tree clouds are a hybrid of trees and tag clouds. To my knowledge, they were first introduced by Jean Véronis. Here is an example:

Philippe Gambette has made available Python software to generate tree clouds. Together with Jean Véronis, he also gave a presentation on the topic recently: