The Information: A History, A Theory, A Flood [Paperback]

I'm not sure the reviews so far are terribly helpful if you want a quick feel for whether to read this book or not. So here goes.

It's basically a bravura sweep through the history of information, told with great panache and lots of anecdote, mixing straight narrative with reflection on wider significance, and attempting to explain quite difficult concepts for the non-specialised reader. Whatever else it may or may not be, I found it a lively and enjoyable read.

The book falls broadly into three sections. The first runs through key early stages in the creation, storage and use of information - the alphabet, printing, the telegraph, telephone, etc. I didn't find much new here but the author did a great job marshalling facts, figures, characters and anecdotes into a lively tale.

The heart of the book grapples with information as a scientific concept, and you will find yourself in the realm of computers, information theory, DNA and quantum mechanics (to name but a few). This isn't natural territory for me, but I was swept along by Gleick's style and even felt I understood some of the underlying mathematical concepts he sought to explain.

The final section is essentially a thought piece on the modern information age, considering the ubiquity of information from the internet and the perils of information overload. Rather like the first section, I didn't feel there was a great deal new here but Gleick's ability to call up literary references, make parallels across the centuries and ask the pertinent questions made it an engaging read. I'm certainly pleased to have made the acquaintance of Vincent of Beauvais, a thirteenth century monk who seems to have arrived 750 years early for the Information Age.

So, a dazzling read certainly, but one also with a great deal of substance. Recommended.

Gleick's Chaos was one of the books from my teen years and I read Genius his biography of Richard Feynman. So I had high expectations for The Information. In chaos Gleick was at his best when he was telling the history of chaos or the biographical sketches of the participants (Chaos focused on Mitchel Feigenbaum). Again here these are the strongest parts of the book when he is talking about Babbage or Shannon. The stories of Shannon and his seminal paper on information theory is brilliantly explained along with the impact of the advances in communication from telegraphy to telephones and the internet. His explanations of probability and complexity are much clearer than similar arguments made by Murray Gell-Mann in the Quark and the Jaguar. He also does a great job of number theory and the problems of rational, irrational and computable numbers and their information content. It was very nice to see Henry Quastler mentioned as he has unfortunately been ignored because of his untimely death.

The problem comes when towards the end in the chapters when he looks at information in biology, entropy and complexity. Biology as Sydney Brenner and Craig Venter have both said is an information science, but Gleick looks too much at the Ricahrd Dawkin's view of neo-Darwinism and information, which is a gloss on the work of John Maynard-Smith. Another founder of the idea of information in biological sequences, especially from a phylogenetic perspective is Linus Pauling. The real truth in biology is deeper than this. The gene code is not a code for an organism the same way as a blueprint does not build a building without the builders to build it and the technical know-how of the construction. We are seeing some progress in understanding the complexity of biology, and it is much more complex than we have ever imagined. This is the field of bioinformatics and bioinformaticians as far as I know (I have been one for nearly 20 years) do not investigate chain letters, which is the only example he gives in the book. The information in the genome is at many different levels, in sequence, in structure, between the genes in the junk DNA and in the chemistry. It is not 0 and 1 it is an intimate layering of information.

I do not think he was trying to be comprehensive and so the work of Steven Wolfram, Gell-Mann and those at Santa Fe like Andreas Wagner as well as older work from Conrad Waddington to provide a broader view of the arguments about complexity in computational and biological systems.

I've been a fan of James Gleick's work ever since his book on Chaos came out. Thus I was looking forward to reading his book on Information. Unfortunately, for the first time I was disappointed. It's difficult to put a finger on a specific reason why I should have only got part way through the book before abandoning it. I'm used to reading larger books than this, so it wasn't the size. The writing seemed more turgid that past work, and the portraits of the key figures more fuzzy. And, of course, it wasn't helped by the tiny print size used by the publishers.

Unlike some of the other reviewers, I don't have any specific disagreements with Gleick's ideas on information, at least as far as I got through the book. It was just that the writing was too pedestrian to hold my attention for long enough to complete it.