# Trevor's Reviews > The Information: A History, a Theory, a Flood

Rate this book
Clear rating

by
's review
Apr 17, 2011

really liked it
bookshelves: science, maths, history

I think this is perhaps as good an introduction to information theory as you are likely to read. Lucid, clear and quite nicely paced, it covers a wealth of material and it does so with beautiful ease. This guy really is a wonderful science writer. His Chaos and Newton were both stunning books. I got about half of the way through Genius, but then got distracted and never quite made it back – but I’ve always meant to. All the same, this one shines and shines.

Perhaps the best chapter was the one on randomness. Randomness is such a tricky concept, but oddly, not something we generally really think about. Just how do you go about proving that a series of numbers is, in fact, random? The problem is that we humans are hopeless at spotting randomness. This is partly because we are such excellent pattern picking machines that we even spot patterns when there are none. And then we also tend to think there must be a pattern if there is repetition, but random sequences have odd repetitions too. All this means we tend to think things that aren’t random are in fact random and vice versa.

The definition of randomness is that there is, in a string of numbers, a one in ten chance that you will be able to pick the next number in the sequence – in a random series you will have a one in ten chance – if you can do better than this chance at picking the number then you must have an algorithm to help you pick the next number and that means the next number can’t be random.

His discussion of Turing, not just his test but also his machine and incalculable numbers, is highly readable and clear. His discussion of Gödel is somewhat less clear, but than I’m yet to have read a perfectly clear description of the incompleteness theorem – which might say more about me than it does about the descriptions I have read, who knows. This one is still good, even if it remains over my head. However, there is a wonderful discussion of the relationship between information and entropy and why entropy is an important concept for people to understand, as good an explanation as any I have ever read.

The early parts of the book are a joy. The stuff about the barbed wire telegraphs is particularly fascinating. As was his explanation of why multistorey buildings needed the telephone to be invented as much as they needed lifts.

I was less impressed by the discussion of memes, but mostly because I don’t find this nearly as useful a metaphor as others do and worry when ideas that are clearly meant to be provocative end up being taken much more seriously than they warrant. Selfish genes and selfish memes, with their characteristic inversion of common sense, tend to become Blackmore’s ‘self-plex’ and the end of human freewill and identity and therefore take a joke all a little too far.

The discussion at the very start of this book about African talking drums is virtually worth the cover price alone. I had never realised that these communicate ‘tonally’ and that to make them work the drummer must add lots of redundancy to the message, almost like a convention of sub-phrases. This was a wonderful description of why redundancy is necessary to messages and said interesting things about Western racism. Westerners simply could not believe these drums actually could send messages or that they were not being sent by a kind of Morse Code.

This book really is a pleasure and on a fascinating topic that is deftly handled by one of the best science writers alive.
flag

message 1: by (new)

Trots off to order this....

Knitters have a thing called a yarn diet. I've been on a book diet and you've just gone and made me break it.

message 2: by (new)

You'll love it, NGE - the early stuff is particularly good. I'd never really thought about the telegraph before, about how hard it must have been to invent, how many bits had to come together. Not just the code, but how to transmit the code. It is fascinating how hard it is too think 'simple'. Just because you can do elaborate things doesn't mean you should.

Anyway, book diets are made to be broken, so I can't feel too bad about getting you to break yours.

message 3: by (new)

Trevor wrote: "You'll love it, NGE - the early stuff is particularly good. I'd never really thought about the telegraph before, about how hard it must have been to invent, how many bits had to come together. No..."

Actually, I'm especially interested by patterns and randomness. I spend a lot of time trying to explain to bridge players that what look like patterns in the hands dealt are random and not patterns at all, but it is very hard to get that across. Trying to struggle with some of these ideas in statistics at the moment. Maybe I will get some illumination from this!

message 4: by (new)

The part of the book on randomness was easily the best. He talks about pi - how it is impossible to guess what the next digit will be (well, unless you calculate it) and so it is both random and not random. I think this book is going to trouble me for a while yet. I really wish I'd paid more attention when I studied statistics decades ago. I wish someone had made it clear to me that statistics wasn't just about gambling, but rather that it was about everything.

message 5: by (new)

Still waiting for the public libraries to process the copies they have bought ...

message 6: by (new)

Trevor wrote: "The part of the book on randomness was easily the best. He talks about pi - how it is impossible to guess what the next digit will be (well, unless you calculate it) and so it is both random and n..."

I totally agree with you. I'm bitterly regretting the same thing. I was wondering if there is a book for lay people who need reliable methods of statistical testing...I've been looking into Fisher and Bonferroni and that's taking me further afield, but I feel like I will never know enough to be able to make my own judgements as to what is best. Meanwhile, the academic world seems to use methods which are often spurious but which they find it easy to use.

This is an example, just in case you are interested:

http://www.plosmedicine.org/article/i...

and a fascinating US Supreme Court document I'm wading through:

http://www.americanbar.org/content/da...

message 7: by (new)

Oh, I must read that, I'll try to do it tomorrow when I'm more awake. I had thought that 'significant' might mean, you know, significant. So that when someone said they had carried out research and it had a significant finding, well... But to find out it means nothing of the kind is very annoying. And then I thought 'effect sizes' might be the way to go with this sort of thing - but they seem quite open to interpretation too, but much more of what I thought significance was supposed to mean. Thanks for these links.

message 8: by (new)

Trevor wrote: "It is fascinating how hard it is too think 'simple'. Just because you can do elaborate things doesn't mean you should."

I recently ran across a statement about the space programs: The USA spent millions inventing pens that would write upside down; Russia gave their astronauts pencils.

message 9: by (new)

The story is actually not true, Lindig - but the real story is even more interesting that the made up one. Which, in itself, is amusing.

message 10: by (last edited Jun 30, 2012 06:19PM) (new)

I love my Fischer Space Pen. However the trait I'd need in a Pen is one that can not be lost or left behind.

message 11: by (new)

That *is* interesting. Did you know that already, or did you check it out just for me?

message 12: by (new)

I think it is mentioned in Packing for Mars: The Curious Science of Life in the Void, but I can't remember now. I had to check and hadn't remembered the full story. I find it amazing when the made up story isn't quite as interesting as the real story. But I guess memes need to be snappy.

message 13: by (new)

Trevor wrote: “This guy really is a wonderful science writer.”

Yes, he really is. Except for Faster, I found his other books outstanding. I still think about his description of the development of communication, language, writing, and number systems.

message 14: by (new)

I haven't read faster - I read the blurb and moved on.