Shannon and Weaver

Revisiting 'Shannon and Weaver', I've realised that my interests develop from Warren Weaver's paper.

It always pays to go back to the early papers in any field, and though I'd recently re-read Shannon's paper, I'd not gone back to Weaver. Weaver talks about three levels to the general communication problem:

• Level A. How accurately can the symbols of communication be transmitted? (The technical problem.)
• Level B. How precisely do the transmitted symbols convey the desired meaning? (The semantic problem.)
• Level C. How effectively does the received meaning affect conduct in the desired way? (The effectiveness problem.)

(Shannon is primarily focussed on Level A.) I suspect that three levels is artificial - you could choose any number - but maybe is useful in practice. (Also of course the number three has religious significance, as in the trinity. You may mock, but I'm sure the OSI model has seven layers because of the religious/mystical significance of the number seven. It is certainly why we've identified seven colours in the rainbow, though it meant using 'indigo'!)

As I was reading, I was thinking 'ah, but he's not explicitly talked about context' and then there it was, a mention of context, exactly fitting with where I was coming from. Unfortunately I can't actually find it again, just at the moment... ("If I had a pound for every time I've seen something and not made a note of the reference..." Well, I was squashed on a crowded train at the time, among the 33 000 people travelling from Milton Keynes to Wembley to watch the MK Dons in the Johnston's Paint Trophy final, so it wasn't terribly convenient to take notes.)

Hirst's £50m skull

All art relies on context.

A fuss about Damien Hirst's diamond-encrusted human skull again, reported in the Independent yesterday:

Yesterday, Ralph Rugoff, director of the Hayward Gallery in London, issued a devastating critique of the work's artistic merit, discounting it as a "decorative object"

Now, I'm on dangerous ground here, I've no expertise in art and no artistic training and I'm not claiming to be able to make any judgement on a work of art, but it seems to me that the value of any art is to large degree to do with context. This is most obvious with conceptual art. Duchamp's urinal takes on its artistic significance from what it is presented as, where it is and what is said about it, as much (or more) than what it is in itself. But I'd like to argue that this is true to some degree of all art. Yes, of course, there is skill and craftsmanship in 'traditional' art, but skill and craftsmanship alone does not create a work of art. The art comes from the understanding of the context that the artist brings to their work, a context shared with the audience. This links closely with what David Weinberger was saying about the 'unsaid' (see my comments in an earlier post). He talked through an example of a poem.

Physics is losing the plot

Physics seems to know less now than it did 20 years ago.

I started reading New Scientist again about a year ago - we subscribed to it to encourage my son who is taking Physics A Level. My perception is that the stories that physics is telling about the world don't have the conviction they did 20-30 years ago. Back then, we had sets of elementary particles in neat symmetry groups The models predicted particles that we'd look for with bigger accelerators. In astrophysics the universe was expanding following the big bang, and we just needed to find out how much matter there is in the universe to find out whether it would expand for ever or whether it would slow and reverse, back to the big crunch.

Today, New Scientist is full of debate about whether string theory really consitutes a valid model since some say it's not testable, and wierd debates about many-universe theories. Then in this week's issue there is debate about dark energy and dark matter, which, we are told, must account for 96% of the universe. We've never detected dark matter, and, listen to this:

The uncomfortable truth is that the more detailed our observations of the universe, the more confusing the dark matter picture becomes. Sometimes there is too much dark matter [...] Other times, we see too little dark matter.

The time must be right for a genuine paradigm shift, in Kuhn's sense.

And, of course, I think that it has got something to do with information.

Wikipedia rage

I was moved to make the following comments recently:

Every time I search the web for something for T325, alkaline batteries, proton exchange membrane fuel cell, wikipedia is always there among the first few results. And it generally is very useful (although rarely sufficient to get the understanding that I want).

It is beginning to annoy me. Wikipedia-rage. (Particularly since I've just read that the originator of wikipedia originally got rich from a search engine designed specifically to locate pornography on the web.)

Actually I think Wikipedia is quite extraordinary.

Some sceptical colleagues have argued recently that there have been lots of over-blown claims for Wikipedia, and you could find equivalent claims having been made about conventional encyclopaedias in the past. That may be true, maybe you can trace a continuum from the great historical encyclopaedias to Wikipedia, but sometimes incremental quantitative changes build up to a qualitative change, or so it seems to me. It is like emergent properties. The old issue of reductionism again.

Here's one feature of Wikipedia that I'm fascinated by. You have to keep working on an entry, otherwise it will deteriorate. There is so much vandalism (some of which is really bizarre - what on earth posses someone to insert the word 'Gay' in the middle of a description of fuel cells?) that unless someone is there to sort it out, the entry, I believe, would sooner or later be worthless. There is a similarity here with a living organism. OK, I know that saying something is like a living organism, calling something 'organic', is a cliché. But I happen to think it says something useful here.