Never Let Go is a collection of Dan John's writings on strength and conditioning training. Subtitled "A Philosophy of Lifting, Living, and Learning", it provides a pragmatic approach to navigating the tangled web of information about health, fitness, and athletic training. A certain amount of strength and conditioning is necessary for good health and is fundamental to fitness and athletic performance. How much and what kind depends on your goals, and Dan emphasizes that you need to need to stay focused on your goals.

The main points of Dan's strength and conditioning philosophy are:

1. The Body is One Piece.
2. There are three kinds of strength training:
   • Putting weight overhead
   • Picking it off the ground
   • Carrying it for time and distance
3. All training is complementary.

Each chapter is a lesson in the guise of an entertaining and enlightening story told by a master practitioner of the strength arts. Reading and rereading these stories will help you in your own strength practice, whether as a coach, athlete, or fitness enthusiast. As you read it you will sometimes think "I knew that", while realizing that you've just reached a better understanding. Dan teaches movements rather than muscles and you need to train the full spectrum of normal movement. However, if you try to emphasize everything, you end up emphasizing nothing, so he recommends a warm up that works a variety of movements which allows you to focus on specific goals in the main workout.

Since the body adapts to novel stresses, everything works but nothing works forever. Combined with everyone's unique genetics and history, this implies that you should try different things long enough to see how they work for you. If they don't work, move on to something else. If they do, continue with them until you stop progressing, then try something else. Keep a training log so you'll know what has worked in the past and can profitably be repeated.

"Never Let Go" is a worthwhile and enjoyable read for just about anyone. It's a great book on the pursuit of althleticism that should be read and re-read by anyone with more than a passing interest in health, fitness, and/or sports performance.

A friend of mine in the disc drive insustry laments the fact that the One Laptop Per Child XO uses flash memory rather than a hard disc for persistent storage. Yet even without disc drives in the laptops themselves, these machines will drive demand for hard discs. While mostly self-contained, the XO is designed to work with school servers of various sizes that will have disc drives. For disc drives to be attractive for an XO successor, they would need to be very small, very rugged, and very cheap. That's a tall order, but I think we can get there.

In The Math Instinct, Keith Devlin looks at mathematics in nature and how abstract math is related to it. Many animals are hard wired to perform various mathematical computations that help them survive. Only humans seem to have extended such natual ability to more general problem solving.

ffnet is a feed forward neural net library for Python. The heavy lifting is done with Fortran and NumPy, so it should execute quickly while allowing the user to model the NN within Python. Version 0.6 has just been released and it looks useful for pattern recognition and machine learning work.

Open Bayes for Python is an Open Bayes implementaion written in Python rather than C++. Bayesian Networks are useful for reasoning under uncertainty and machine learning. Open Bayes for Python uses Numarray for the heavy lifting, but the project is moving to Numpy for future development.

Since baesian networks are computationally intensive and trivially parallelizable, it will benefit from the IPython1 work.