Sir Charles Darwin speaks at VUC

His first lecture, despite a recommendation by the censor for the most learned only, was an intriguing account of the still speculative theories about the constitution of the earth.

The Memorial Lecture came next, a description of Rutherford's work in establishing the "planetary" picture of the atom, and hence atomic number. It was particularly interesting because Sir Charles had been on the spot and actually taken part in some of the research—he was one of Rutherford's original team at Manchester.

However, it has been reported elsewhere, it would be better to attempt to give some account of the last of his three lectures. This was an introduction to a fascinating and extremely modern subject called the "Theory of Information."

There are several examples in physics of what seems at first sight to be a rather unexpected limitation on the amount of information which can be passed through a given system. For ex-sample, if we use a telescope to obtain information about a star, then no matter how strong the lenses we use, we find that there is an upper limit to the detail we can ever observe, which is fixed merely by the diameter of the first lens (after a certain point, using stronger lenses makes the image larger, but much fuzzier).

Similar phenomena occur in radar and in telegraphy. They are of course particularly vital in telegraphy, and led theoretical telegraphists to ask such questions as to whether there is a limit to the information which could be transferred even through an ideal system; what we might mean by information anyway, and how can it be measured and communicated?

(Although Sir Charles spoke mainly with the application to telegraphy in mind, the answers to such questions reach extremely diverse fields, from biology to electronic computers, and he mentioned at the end of his lecture how they might be related to the way in which our own minds work.)

From the telegraphist's point of view the English language is far from efficient. It is redundant—about 50 per cent. of the letters on a page could be blotted out without seriously impairing the information garnered by the reader. It is a most uneconomical code.

However, there is a very important problem in which redundancy is useful. This is the problem of noise—the random occurrence of spurious signals (like static on a radio) which interfere with the proper signal. Redundancy here is a safeguard; it ensures that enough of the right signals are received for no information to be lost.

The actual percentage of redundancy can be calculated which would be needed to counteract a given amount of noise.

Perhaps this at least makes plausible a feature of the theory that is rather remarkable: that although it seems very abstract, with its feet far from firmly planted on the ground, it has chiefly been developed by, and proved most useful to, hard-headed and practical business firms, such as the telephone companies in America._