34

Two Letters

The script for 'The Sea in their Blood', my next film, is nearly finished. As you suggested, I consulted the National Office of Statstics' 500-page official handbook to Britain. Did you know that biscuit exports were valued at 287 million pounds or that the Average Person watched more than 25 hours' televison a week and listened to the radio for 10 hours? The last two facts may explain why cats, who do not need walkies, outnumber Britain's 6.6 million dogs by more than half a million. I notice that British airlines carried 77 million passengers and more than 772 million journeys were made on London Underground trains last year. Unfortunately, the handbook does not state how many reached their destinations.

470,000 tonnes of fish were consumed last year along with 9.48 billion eggs and 28 billion pounds worth of alcohol. Who ate and drank all this remains disputed. Possibly men, whose life expectancy ended abruptly at 74. Or their widows, who had another six years to drink away their loneliness before expiring. Of those who survived last year, three quarters had the pleasure of a microwave, 82 per cent had a video and 91 per cent a washing machine. The richest 10 per cent of the population has 48 per cent of the wealth. Each day in England and Wales, 700,000 people visit their family doctor, 130,000 attend the dentist and 2,000 babies are born.

The FO is keen for me to finish the film as soon as possible since visitors to Britain have more than doubled in 20 years to 25.3 million.

Yours,

Peter.
Hammersmith, London.

I've been reading an interesting article on Bacon Numbers in the Guardian newspaper and, although it has nothing to do with your work for Session Three, I thought you might like a summary of it.

Bacon Numbers are a measure of how closely an actor/actress is related to Kevin Bacon (the Hollywood star) in terms of co-appearances. You calculate an actor's Bacon Number like this. Bacon himself is assigned the number 0. Anyone who has appeared in a movie with Bacon gets Bacon Number 1. (Tom Hanks and Bacon both starred in Apollo 13, so Hanks has Bacon Number 1.)

Next you look at any actor who has not appeared in the same movie as Bacon, but has appeared in a movie with someone who has. All these people have Bacon Number 2. (Sally Field has Bacon number 2 since she appeared with *Tom Hanks* in 'Forrest Gump' and has not appeared in the same movie as Bacon.)

An actor has Bacon Number 3 if they were in a movie with someone who was in a movie with someone who was in a movie with Bacon, but there is no shorter movie path to Bacon. And so on.

Of more than 300,000 actors listed in the Internet Movie Database, 263,484 can be linked to Bacon in this way. Of these, 1,267 have Bacon Number 1 78,867 have Bacon Number 2; and 149,018 have Bacon Number 3. After that, things start to drop off. 32,094 have Bacon Number 4 1,903 have Bacon Number 5, and 299 come in with Bacon Number 6. A mere 34 have Bacon Number 7 and just 1 has Bacon Number 8.

I wonder what a film - in the manner of THE FALLS - of those 34 people who have Bacon Number 7 would look like?!

The Bacon list is at http://www.cs.virginia.edu/~bct7m/bacon.html

Apparently this method of data mining is being used by supermarkets. What, for instance, goes into the same shopping trolley as sausage meat (and thus has sausage number 1), and if they buy those other things, what else might they buy (items having sausage number 2)? When they know that pattern, they can use it to arrange the goods in their store to increase sales.

If you've got this far in this letter then you might be interested to know that 'Numbers: The Universal Language' by Denis Guedj got a good write-up in the Guardian because it's 'packed with numberly fun'.

Best,

Peter,
Hammersmith, London

.........

Peter Greenaway

.........