I am a bit surprised to see this on HN without context. This was originally an email from Jeff to our mailing list and was a pretty good overview so we dumped it into a wiki page. For those new to the problem, the page summarizes it well:

> The basic idea of temporal pooling is patterns that occur adjacent in time probably have a common underlying cause and therefore the brain forms a stable representation for a series of input patterns.

While the idea is simple, the implementation is tricky. We recently made our experimentation public here: https://github.com/numenta/nupic.research

Jeff also had a really informal discussion about the topic at one of our hackathons: http://numenta.org/blog/2014/05/09/2014-spring-hackathon-out...