Tag Archives: Computing

Network Geeks – Brian E. Carpenter ***

Posted on by
Reply

There is a series of TV adverts in the UK that have managed to embed their tagline into common usage. The ads are for a type of varnish, and that tagline is ‘It does what it says Screenshot_13_05_2013_10_05on the tin.’ There is a real problem when a book doesn’t do what it says on the tin – you get cognitive dissonance, expecting one thing and discovering another. That’s what happened when I opened up Network Geeks.

The subtitle promises ‘how they built the internet.’ Now this is a topic I’m fascinated by. I really enjoyed the book Where Wizards Stay Up Late, which details the story of the origins of the internet, but that’s quite old now, and I assumed this would give a modern day take from the viewpoint of an internet dominated society. What you get inside is totally different, and that’s a shock.

In trendy music terms, this book is a mashup. It really has three separate themes, only linked by the author, Brian Carpenter. One is an autobiography – so we get a fair amount of Carpenter’s family history, going back a good few generations. It’s not badly written, but probably of limited interest to anyone outside Carpenter’s family. Secondly – and this is the best bit – we have a considerable account of Carpenter’s work at CERN. He worked there twice and if you are into the developed of distributed computing (as I am) there is some really interesting material here, as CERN was both groundbreaking and yet isolated from the mainstream. Apart from anything else in this technical memoir part of the book I had distinct tugs of nostalgia as I had a great time working on DEC equipment, which regularly rears its head, while in the OR department of British Airways.

So far, so good – but we are yet to encounter anything that really has to do with the supposed topic of the book. This comes into the third part of the mashup, featured in the introductory section (which is part of the reason it is such a shock when the book suddenly goes into autobiographical mode) and towards the end. But this isn’t really about ‘how the built the Internet’ at all. It is about ‘how their committees made endless bureaucratic decisions about the architecture and protocols of the internet and how the architecture and protocols developed.’ To be honest, that is a rather less exciting, and certainly a lot more specialist field.

The problem is, unless you are really into the nitty gritty of how the committees that control the internet work, this probably isn’t for you. Carpenter falls into a few writing traps in naming far too many people we aren’t really interested in, using endless acronyms we don’t really care about and giving much too much detail on the minutiae to the extent that we lose the big picture. Here’s a not atypical snippet to get a feel: ‘Internet standards, originally endorsed by DARPA, came from the IETF by 1991, and certainly not from the ITU or the ISO, the twin homes of CLNP. On the other hand, CLNP was officially defined and had already been picked up for the next version of DECnet, a significant factor in the minicomputer market then served by the Internet.’

It’s not that this is a bad book – it just doesn’t do what it says on the tin, and I can only recommend it for the rather narrow audience for whom this kind of thing is meat and drink.

Paperback:  

Kindle:  

Review by Brian Clegg

The Golden Ticket: P, NP, and the Search for the Impossible – Lance Fortnow ***

Posted on by
Reply

There is good and bad news early on in this book about the P versus NP problem that haunts computing. The good news is that on the description I expected this to be a dull, heavy Screenshot_04_05_2013_11_02going book, and it’s not at all. Lance Fortnow makes what could be a fairly impenetrable and technical maths/computing issue light and accessible.

The bad news is that frustratingly he doesn’t actually tell you what P and NP mean for a long time, just gives rather sideways definitions of the problem along the lines of ‘P refers to the problems we can solve quickly using computers. NP refers to the problems to which we would like to find the best solution’, and also that he makes a couple of major errors early on, which make it difficult to be one hundred percent confident about the rest of the book.

The errors come in a section where he imagines a future where P=NP has been proved. This would mean you could write an algorithm to very efficiently match things and select from data. Fortnow suggests that our lives would be transformed. This is slightly cringe-making as fictional future histories often are, but the real problem is that he tells us that the algorithm would make it possible to do two things that I think just aren’t true.

First he says that from DNA you would be able to identify what a person looks like and their personality. Unfortunately, these are both strongly influenced by epigenetic/environmental issues. Anyone who knows adult identical twins (with the same basic DNA) will know that they can look quite different and certainly have very different personalities. And they will usually have been brought up in the same environment. Fortnow is forgetting one of the oldest essentials of computing – it doesn’t matter how good your algorithm is, GIGO – garbage in; garbage out.

The other, arguably worse error is that he says that it will be possible to have accurate weather forecasts going forward X days. This is so horribly wrong. He should have read my book Dice World. The reason you can’t predict the weather at all beyond about 10 days is nothing to do with the quality of the model/algorithm, it is because the system is chaotic. Firstly we just don’t know, and never can know, the initial conditions to enough decimal places not to deviate from the real world. When Lorenz first discovered chaos it was because he entered the starting values in his model to 4 decimal places rather than the 6 to which the model actually worked. It soon deviated from the previous run. We can’t measure things accurately enough. The other problem is that the weather system is so complex – hence the slightly misleading title of Lorenz’s famous paper Does the flap of a butterfly’s wings in Brazil set off a tornado in Texas? – that we can’t possible take into account enough inputs to ever have so good a model as to go forwards that far. Sorry, Lance, it ain’t going to happen.

For the rest, the first half or so of the book goes along pretty well, gradually opening up the nature of P and NP, the problems that are of interest and the ‘hardest’ NP complete problems. I found the main example, used throughout, a hypothetical world called Frenemy where everyone is either a friend or enemy of everyone else confusing and not particularly useful, but Fortnow gets plenty of good stuff in. After that it’s as if he rather runs out of material and it gets a bit repetitious or has rather tangential chapters.

Overall, despite the flaws, a much better and more readable book than I thought it was going to be – but probably best for maths/computing buffs rather than the general popular science audience.

Hardback:  

Kindle:  

Review by Brian Clegg

The Logician and the Engineer – Paul J. Nahin ***

Posted on by
Reply

For its target readership this is an excellent book – and I have to say as someone outside that market I really enjoyed some parts – but the fact remains it is aimed at a pretty narrow segment. There’s even a little section at the front of the book that effectively says ‘read this to see if you can cope with the rest.’

The bits I found particularly appealing were a few introductory logic problems (though I’m not sure I agreed with all  the conclusions) and the pocket biographies of mathematician George Boole and information engineer Claude Shannon. However, while technically qualified to deal with the other parts of the book, in truth I couldn’t be bothered – it was too much like hard work.

For bits of it I would have to wade through far too much grunt maths, and for other bits would have had to think far too hard about electronic circuits and the logic circuits beloved of basement dwellers on computer science courses. (Or was it just my university that confined the computer scientists to the basement?)

I think the author makes the mistake that many academics make when trying to write for a broader audience: they carry through too much of the textbook, and find that the aspects that often encourage people to remember things in that context (often because they involve repetitious grunt work) actually prevent popular science readers from getting the message. It’s a shame, because the subjects are interesting, but unless you are the kind of person who designs logic circuits for fun, this is probably not the book you’d want to see.

Hardback:  

Kindle:  

Review by Brian Clegg

Turing: Pioneer of the Information Age – B. Jack Copeland *****

Posted on by
Reply

Alan Turing is a name that has grown in stature over the years. When I first got interested in computers all you really heard about was the Turing test – the idea of testing if a computer could think by having a conversation by teletype and seeing if you could tell if there was a computer or a human at the other end. Then came the revelations of the amazing code breaking work at Bletchley Park. Now, though, we know that Turing was much more than this, the single person who most deserves to be called the father of the computer (we allow Babbage to be grandfather).

All this and much more comes through in B. Jack Copeland’s superb biography of Turing. It’s not surprising this book (and its competitors) is on sale now. 2012 is the hundredth anniversary of Turing’s birth. And it is a timely reminder of just how important Turing was to the development of the the technology that is at the heart of much of our everyday lives (including the iPad I’m typing this on today).

If I had to find fault at all with this book, it can be a little summary in some aspects of Turing’s private life – but I suspect this reflects the lack of information from a very private man. However if, like me, you’re a bit of a computer geek it would be impossible not to be fascinated by the description of his ideas and the technology that was developed from them, beautifully written by Copeland. I’ve read plenty before about Enigma, but the section on this was still interesting, and the Tunny material (a later, more sophisticated German coding device, to crack which the Colossus computer was developed) was all new to me.

Similarly, I hadn’t realised how many firsts belong in the UK rather than the US. I knew Turing’s work led to the first stored program electronic computer – the first true computer in a modern sense – but I hadn’t realised, for instance that Turing was the first to write the code for computer generated music, with the first computer music in the world produced using that code in Manchester (contrary to the myths you are likely to see online).

Although some of the personal life information is a little sketchy, Copeland really delivers on Turing’s death. I had always accepted the story that he committed suicide with a poisoned apple as a result of the ‘chemical castration’ he chose as an alternative to prison for admitting homosexual acts. Copeland tears this myth to pieces. Turing had endured the hormone treatment with amusement – and it had finished a year before his death. By then he was fully recovered. He appears to have been happy and positive at the time of his death. He left a part-eaten apple by his bed every night. And he was experimenting on electroplating in a room adjacent to his bedroom – using a solution that gave off hydrogen cyanide. The postmortem was very poor, without testing whether the cyanide that killed him had been ingested or inhaled. The evidence seems strong that Turing’s death was an unfortunate accident, not the tragic suicide that is usually portrayed.

In the end I can strongly recommend that anyone with an interest in computing should rush out and buy a copy of this book. Well written, fascinating and overthrowing a number of myths, it’s a must-have.

Hardback:  

Review by Brian Clegg

The Universal Machine – Ian Watson ****

Posted on by
2

I have to admit upfront that I’m a big computer geek and lap up anything about the history of computing, so this could be a slightly biassed review - but even if you don’t have a lot interest in computers per se, their influence on the modern world is so huge that this has to be a book you will at least consider. Because The Universal Machine follows the development of computers, as it says in the subtitle, ‘From the dawn of computing to digital consciousness.’

Ian Watson takes us on this journey with a charming if slightly amateurish personal style (the first line of the book is ‘Hi, you probably don’t know me, but assuming you stick with this book then we’re going to be spending quite a bit of time together.’) – don’t be put off by the introduction, the style settles down. Bearing in mind my bias, I found it absolutely fascinating, from one of the best section’s on Babbage’s work I’ve ever read, through the development of the electronic computer, into PCs and the web.

On the whole, the historical content was at just the right level – enough to keep you interested without getting overwhelmed. I was slightly surprised Ted Nelson, who devised the hypertext concept, wasn’t mentioned, but there is always going to be something. Of course, there are parts of the story where a lot more depth is truly fascinating – so I’d recommend, for example, the classic Insanely Great and Hard Drive on the origins of Apple and Microsoft respectively – but for an overview this was hard to beat. Interestingly it’s at it’s best in an application context. The only time the text got a little dull was when Watson talked pure computing, and in the probably unnecessary future gazing bit at the end.

What was odd about the book is that I know that it was published by a large international publisher (Springer Verlag), but reading it sometimes felt more like the experience you typically have with a self-published book. The text is too tightly crammed on the page, making it slightly uncomfortable to read. And there are rather more typos and basic errors than I’d expect in a professionally published book. One example that jumped off the page – Mary Shelley’s surname is written as ‘Shelly’ at least three times. (Funnily enough, her surname was actually Godwin when she wrote Frankenstein, though she had married Shelley by the time it was published.) There are a couple of places where the text doesn’t knit well together, resulting in some repetition. And it’s also rather unfortunate that the author says ‘you may be reading [this book] on a Kindle or iPad’ when there isn’t a Kindle edition.

Overall, then, although I would recommend following it up with some more focused computing histories, The Universal Machine is a great way to get a real feel for where the machines that are at the centre of so many of our lives came from.

Paperback:  

Review by Brian Clegg

Natural Computing – Dennis Shasha & Cathy Lazere ****

Posted on by
Reply

Here we have a touch of brilliance; an exploration of computing on the edge. What the authors cover very engagingly is the different ways computer can develop, whether through the ‘natural’ route suggested by the title – using bacteria to compute with, for instance – or programming robots to be more like insects than a conventional rational individual. We see software being developed in evolutionary fashion and the attempts to harness quantum computers – reflecting on their capabilities and limitations.

It’s all very readable, though because the book is split into 14 chapters, each based on one or more individuals and their work, I found the biographies that started each chapter a little tedious because, frankly I wasn’t very interested in these people. That didn’t stop their work being fascinating, and I know popular science thrives on context, but this was unnecessary information.

The other slight hesitation I have about the book is that the authors are relentlessly enthusiastic about the outcomes – there could be more examination of chances of success. To take an example, the chapter on Jake Loveless and Amrut Baharambe looks at using evolutionary code to model a financial market and make successful trades. It says at the end that their genetic algorithm ‘worked’ – but what does this mean? Did it do better than random selection? Will it generally? All the evidence that markets really aren’t suitable for modelling and nothing can forecast crashes because they aren’t logical or following any kind of rule (other than occasional panic) – but there was no examination of how this problem was dealt with or why, if this algorithm ‘works’ it isn’t generating billionaires all over the place.

There were several other places where the enthusiasm rather plastered over what could be lack of real results, and it would have been nice to have been able to hold this work up against a more objective measure – but even so it is hugely fascinating for anyone with an interest in computing and how it can continue to change our world.

Paperback:  

Review by Peter Spitz

Donald Mitchie: on Machine Intelligence, Biology & More – Ashwin Srinivasan (ed.) ****

Posted on by
Reply

This is an eclectic collection of writings by and about Donald Michie, the Scottish-born scientist whose career spanned over half a century and covered many topics, most notably computer science and reproductive biology. Michie died in a car accident in 2007, aged 84, and “Machine Intelligence” is a tribute to his life and work compiled by the eminent computer scientist Ashwin Srinivasan.

The book varies widely in style and subject matter, but it is interesting and readable throughout. It comes in three parts, “Machine Intelligence,” “Biology,” and “Science and Society.” Each section is divided into chapters containing 3-5 pieces, with helpful introductions to the chapters by Srinivasan.

The writing is aimed at the non-specialist reader, and specialists may be disappointed by the absence of any of Michie’s many ground-breaking scientific papers. The upside is that experts and novices alike are treated to insider accounts of Michie’s code-breaking at Bletchley Park during WWII, reflections by Michie on how scientists work and the role of government in science, and thoughtful discussions of big topics in AI – such as the Turing test and the role of subconscious or “inarticulate” thought in cognition. Especially worthwhile are Michie’s thoughts on the difference between brute-force solutions to computing problems and truly intelligent solutions.

Michie was much more than a scientist, and some of the most witty and enjoyable writing in the book sees Michie as science administrator, social commentator, and popular science writer. Some of my favourites are his cutting comments on the Lighthill Report (the government report in the early 1970s that almost killed Britain’s nascent AI industry), his article about the reading habits of scientists (they do surprisingly little), and his account of a bizarre trek from London to Moscow that Michie undertook at the height of the Cold War.

“Machine Intelligence” is not a detailed or systematic treatment of Michie’s ideas – it’s a series of snapshots rather than a portrait. Articles on the same theme (like the difference between clever and intelligent computers) are sometimes scattered through the book rather than grouped together. And there are too many typographical errors. But “Machine Intelligence” succeeds as a readable tribute to a remarkable man, giving many glimpses of Michie’s insight, humour, and wide-ranging enthusiasm for science.

Hardback:  

Review by Michael Bycroft

Complexity: a guided tour – Melanie Mitchell ****

Posted on by
Reply

This book made me want to cheer, because with this title OUP has got it right. I dearly love Oxford University Press, and time after time they come up with popular science books that sound really interesting. Only, when you read them they can be dull and not very well written, often, I’m afraid, because the author is an academic. But this time, in this fascinating guide to complexity, emergent systems, networks and more, they’ve found an author with just the right tone who has the ability to make the subject interesting while still conveying her own interest and involvement in the field.

You may have come across complexity as an adjunct to chaos theory – and chaos is covered in here, but there are so many other things too. In looking at the background, Melanie Mitchell includes the theory of information and computation, plus tying this theory into evolution. She introduces us to genetic algorithms and other computer-based mechanisms for systems to evolve, including the potential for using these approaches in problem solving. We discover cellular automata and an attempt to get computers to understand analogy. And there’s a whole section on the hot topic of networks, from the World Wide Web to the human brain. Time and again we see how simple rules and structures can evolve into complex results that can be difficult to predict in their real world forms.

If I’m picky, Mitchell does occasionally give us too much detail, falling into the ‘boring lists’ trap – and some of the items she covers are presented in too technical a way. There’s also a statement at one point ‘Given a room full of air, at a given instant in time each molecule has a certain position and velocity,’ that would have a physicist cringing – for quantum particles, there aren’t values for the properties until a measurement is taken, and even then the uncertainty principle ensures we can’t know both with any accuracy. But the statement is made in the context of some classical statistical physics, so is almost forgivable.

The reader is probably left with a slight sense of doubt. There seems to be a lot of science here that’s fascinating, but can’t really be used for anything. But that’s not the author’s fault, it just reflects the nature of complexity – at least in our present level of understanding – and Melanie Mitchell’s book will certainly ensure that the reader has a good picture of what it’s all about.

Hardback:  

Review by Brian Clegg

Almost Human: Making Robots Think – Lee Gutkind ****

Posted on by
Reply

In my youth I was very fond of business biographies, particularly the ones about the early personal computing world – Apple, Microsoft and the like. I was really inspired by the stories of all those young people, prepared to sleep under their desk so they could get back to the code, or to get the hardware right, burning to make something exciting. I had been a programmer, and I understood this feeling.

I don’t know if it’s because I’m older, or because the world has changed, but I find it difficult now not to be slightly cynical when Lee Gutkind gives us a similar heroic presentation of the all-in working of the young postgrads building robots at Carnegie Mellon University, the focus of his book on the state of robotics. It’s a kind of fly-on-the-wall documentary book – Gutkind spent a lot of time with them – and he’s obviously a bit of a fan. Perhaps part of the reason for the cynicism is that in those early days of Apple, Microsoft etc. the bosses were the same as the workers, while in the university context there’s inevitably a feeling of the older professor throwing the troops to the lions (if you’ll forgive a mixed metaphor), using cheap student labour (making up courses to fit round the things they want to build) because they can’t run to proper research grants.

The other thing that is frustrating to a reader with a business background is the shambolic nature of the operation. People moan about commercial software, but on the whole it works because it’s well planned and well tested – this stuff seems to be neither. There’s too much of an “invented here” syndrome. When I worked for a large company we tried to do a project in cooperation with a university computer science department. We needed some dumb terminals (this was before PCs were common) for the job. Their attitude was “first task is to build the terminal.” Ours was “we’ll buy a terminal off the shelf.” Brought up on one-offs and specials, they couldn’t understand the need to use standard technology – or the benefits in terms of reliability and time saving of using something off the shelf. While there is some off the shelf work in Almost Human, there is still that “build it from scratch” mentality.

The only sense this is a criticism of the book is that Gutkind could be more critical of his subjects – otherwise, it’s a great read. It’s often a page turner as you wait to discover what happened next (though on the whole the answer is the same: the robot broke), and Gutkind gives a great insight into the work of the roboticists, the state of robotics, the interface between the roboticists and scientists, and also the self feeding nature of academia, with two different groups spending all their time on a study of how the others do their work.

It’s certainly an eye-opener if you think the sort of indistinguishable-from-human robots we see on TV and in the movies are anywhere near possible. Just getting a vehicle to go for a drive across open desert on its own is fraught with problems. It’s fascinating and frustrating in equal measures, giving an excellent insight into the state of robot research Carnegie Mellon style.

Paperback:  

Review by Brian Clegg

Colossus – B. Jack Copeland et al ***

Posted on by
Reply

Once upon a time the work of Bletchley Park was so secret that the British Government pretended it didn’t exist. Even after the end of the war this pretence went on, and it is only thanks to recent declassification that the full story is possible to tell. Well, actually, even now it’s not possible – because with what the history of science must surely regard as vandalism, practically everything from the trail blazing Colossus computers to the actual decoded messages were destroyed in a fit of security consciousness that comes worryingly close to paranoia.

A fair amount has already been written about the work on cracking the codes of the German Enigma machines – what is only newly revealed is the attack on the more complex Tunny machines, and the details of the computing factory at Bletchley Park, where ten Colossus computers were eventually deployed against the German codes.

To anyone interested in the history of codebreaking, or the very early days of computing, this book is going to be absolutely fascinating. It’s less easy to recommend it as a general popular science book. It consists of a series of essays by different authors, so has limited consistency of style and readability. And though some of the essays, including that by the kingpin of the Colossus, Thomas Flowers, are readable and as much populated with the people of Bletchley Park as the codebreaking task, some of the others are a little too dense and detailed for the general reader.

It’s a big book – well over 400 pages, and without doubt for the enthusiasts this deserves the full five stars. As a general read it can only really justify three, but even so, it’s well worth taking a look and seeing if it takes your fancy, as it did ours – this was a crucial programme both in the winning of the Second World War and in the development of computing, and it deserves celebrating in a way that the mean minded shutter of supposed national interest has made it impossible to do before.

Hardback:  

Review by Peter Spitz