Sunday, January 08, 2006

Two decent articles from the BBC - shock and horror and pain

What is going on with the BBC? They're coming out with at least two decent articles a week in the Magazine section of their Online News site. That's slightly more than even I manage! Two non-recent examples [why non-recent? Why do you think? Look up the term "lackadaisical" in your nearest and cheapest medical dictionary] really impressed me, to such an extent that I couldn't even talk about them immediately after reading them, or even a week later. No matter: no time like the present. [Invocation of a popular saying, never mind how self-serving, should always be accepted with a graceful "touché".] The first example of this duo is "Risky Business", a history of prophecies or predictions, depending on which epoch you are living in. It's written by a historian (David Cannadine), which probably explains much of why it's such an enjoyable and informative it, even if its length is disappointingly short. The piece climaxes (with good timing) by answering, or at least attempting to answer, the question "What is the point of even trying to predict the future?" The answer is very convincing:

This government, or any other, has to have a policy on motorways, on pensions and on global warming - all of them ultimately based on little more than educated guesses as to what might happen at some far off date when most of today's politicians will long since have vanished from the scene, and won't have to take responsibility for their decisions.
Very true, sir. How do you propose we proceed then?
The way to answer this is to use something that's called the Turnbull Matrix, named after the civil servant who invented it, to try to establish the balance between the likelihood of something unpleasant happening, and the impact if it did. A nuclear explosion, for instance, is relatively low likelihood (we hope), but very high impact (we fear), and there are many, many other permutations.
Turnbull Matrix? What the hell is that? According to this cursory description, it sounds like a basic implementation of Decision Theory, with the principle of taking actions that maximise expected utility under conditions of uncertainty. A google search, though, turns up only... seven results, the first of which is the BBC article, and the others are connected to someone called Brennan Wilcox. Very bizarre. Apart from this detail, I do recommend you read this incisive work. Have I hyped it up enough yet? The second example of the BBC actually publishing something educational and well-written involves the first example I've ever seen of proper maths on there. There's even a formula. All this just to answer the question "Do you get less wet if you run in the rain?". There's even an eminently understandable (to me, anyway) explanation of how to solve this problem systematically, by approximating in a very clear and justified way, and then, when the answer has been found from that, interpreting it in the light of the assumptions, and thus tweaking it. Brilliant. I'm going to watch out for more publications by the author, Nick Allen, who
is a Master of Science in astrophysics and a fellow of the Royal Astronomical Society. An entrepreneur and inventor, he was also co-developer of MouseCage, a disability software www.mousecage.org. Nick continues to teach physics to advanced students at Valentine's High School, Ilford.
Lucky students. Enough already.

1 Comments:

Anonymous Anonymous said...

That's really kind - thanks!

Nick Allen

4/03/2006 12:41:00 am  

Post a Comment

<< Home