Welcome back, fellow readers! This month we read The Man Who Knew Too Much: Alan Turing and the Invention of the Computer by David Leavitt. Join me in the thread to discuss this book and talk about Alan Turing. Also, below the jump is this month’s book-themed recipe (savory scones) and a cocktail recipe provided by Anne Sauer at Mad Art Lab. At the bottom of this post is details for next month’s book and meeting date.
If you’ve never been to a book club before, the rules are simple: Read the book, discuss the book, and keep the discussion related to the book. Or if you haven’t read the book but you’re fairly knowledgeable about the subject, feel free to participate.
I did not really know too much about Alan Turing other than a few facts here and there about the Turing Test and his punishment for being homosexual. This book talked about his background as a teenager and his life at various universities before he signed on to be a codebreaker for the British Government. It was there that he and his team cracked the Enigma code used by the Germans. The parts of the book that were about Alan Turing’s personal life were interesting to read about. However, the book was dominated by large sections explaining the basics of Turing Machines, programming, and math theory, which I found very dry and hard to read. However, I’m just a biologist, maybe if you’re into computer science or mathematics you enjoyed those sections more than I did? Also, I’d like to hear your thoughts on the possible diagnosis of Asperger’s Syndrome for Turing, which is not really discussed in the book.
This month’s themed recipe is: Sun-dried Tomato and Feta Cheese Scones. I chose it because scones are British and the ingredients are non-traditional, in the same way that Alan Turing was British and many of his behaviors were unexpected for the time. (I know it’s a reach, but there’s only so much I can do with food! Also, I definitely didn’t want to do an apple recipe.)
I adapted the recipe from this website, and I only changed a few things: I used diced sun-dried tomatoes in oil instead of roasted tomatoes (I was in a rush), chives instead of green onions, and I had to bake mine for an additional 10-15 minutes (until the outside turned slightly past golden brown). Also, the recipe recommends that you put the ball of dough on a floured surface, but I just went ahead and put it on the parchment paper on my baking sheet. Everyone loved it and at the end of the meeting I only had 1 left!
Photo source: Anne Sauer
Also this month we have a cocktail recipe called the Turing Test, because as Anne says, “You’d have to be a robot not to like it!”
Turing Test
1.5 oz English gin
1 oz honey syrup*
1/2 oz fresh lemon juice
2 dashes Angostura bitters
Strip of lemon peel
Combine gin, honey syrup, lemon juice, and bitters in a cocktail shaker.
Add a generous amount of ice and shake vigorously until chilled.
Double strain into a cocktail glass.
Squeeze the lemon peel over the drink to express the oils, run it along the inside of the rim and then set in the drink to garnish.
*To make honey syrup, combine equal parts honey and hot water (not boiling–hot from the tap is fine) and stir until well-integrated.
——————————
Details for next month: The Skepchick Book Club will read The Last Greatest Magician in the World: Howard Thurston versus Houdini & the Battles of the American Wizard by Jim Steinmeyer, and we’ll meet on July 29th at 11 am.
So, start your Turing Machines, fire up your oven, get out your cocktail shaker, and then join us in the comments!
This is random, but thank you for not going down the apple route with the recipes. I have to admit I find it a bit distasteful that the item Turing most likely used to kill himself is used to represent him so much, even if it is heavily symbolic and otherwise innocuous. I’ve been planning a blog post on it for ages, as I’m apt to do.
Also, FIRST!!1!11! or whatever internet people do when they get to a comment thread early.
The man did so much interesting stuff, I didn’t want to dishonor him by making a recipe themed around his death.
That cocktail looks delicious! I will definitely have to try making one.
I’m an electrical engineer and I do have a background in math and computer science, so I rather enjoyed the technical descriptions in this book. (Yay algorithms!) But even as I was reading them, I was wondering whether those sections would be interesting to readers without a math/CS background.
Turing’s story always breaks my heart. The chemical castration is obviously horrifying and I don’t want to downplay that, but I’m always struck by the sheer stupidity of not allowing him to continue his technical work. As the book kept repeating,
Turing believes machines think
Turing lies with men
Therefore machines do not think
Once you find out that your heroic code-breaker prefers men to women, he’s not only humiliated and despised, but even his technical work is made suspect, and he certainly can no longer be trusted with government secrets. It’s just so idiotic.
I don’t have the right background to really know how likely the Asperger’s diagnosis is. I will say that I do think a high level of social awkwardness (without necessarily being on the spectrum) combined with being homosexual in an age when it was criminal would make anyone be, well, odd, at the least.
By the way, anyone who really enjoyed this book who hasn’t yet read Neal Stephenson’s Cryptonomicon should really do so. You’ll love it. Also, to Mary, nice choice on the next book! I’ve had that one on my Amazon wish list for a while, and I’m glad to have a reason to buy it.
I read a paper a while back that put forth a pretty convincing case for Turing being on the autistic spectrum. I have a copy on my computer but it seems to be no longer available online. Here’s the citation anyway:
O’Connell, H; Fitzgerald, M (2003), “Did Alan Turing have Asperger’s syndrome?”, Irish Journal of Psychological Medicine (Irish Institute of Psychological Medicine) 20: 28–31, ISSN 0790-9667
At the same time, this is all hypothetical – there are all sorts of issues with retroactively diagnosing someone.
Seconding “Cryptonomicon”! I have only gotten a third into it, but it’s great so far. Not one for light holiday reading though…
I was really haunted by Turing’s poem too. And did you see the section where he finally got into the Royal Society and Jefferson wrote him this nasty note about how his computer parts are probably registering this moment as happiness?
Oh, and timely reference here: During one of their last debates, Jefferson basically said that he would not believe a machine could have intelligence until it sexually harassed a lady machine. The more things change, I guess.
I have heard whispers that Turing was autistic, and someone mentioned that it would explain the obsession with Snow White (because people with Asperger’s tend to develop obsessions with random things).
One of the things I admired about him was his refusal to hate himself for being homosexual. And his eccentricities were interesting, but then again I went to an engineering school so I’m not too surprised by his behavior.
Oh, and I’m glad you like the new book, I’ve had that one on my list for a while too and it seems like a fun summer read!
I think speculating about historical figures’ possible psych diagnoses is comparable to discussing TV characters’ astrological signs.
It can be an entertaining diversion, but nobody should take it too seriously. And it can have the destructive side effect of feeding into stereotypes and misconceptions.
Ironic that this book is not available on the Kindle. Only in Hardcover or Paperback.
Brian Bowman
I know, I was thinking that too.
Huh. Always math. Drives me nuts. I am pretty good at programming, but not *that* great at math, having missed at least one grade in it, and, more to the point, almost never using it when writing programs, unless its something specific, I can look up, or a library, which someone else already made.
But, I can see why they felt the need to go into all the theoreticals, and computations. Its just not necessary to explain what programming *is*, or how to do it. Somewhat sadly, its not even necessary any more to know the basics of how the machine works, in most cases, or how binary works, to code. I have, more than a few times, had people look at me funny because I used binary math to do some simple calculations, instead of an if/then. This is a habit/hold over, which may or may not mean anything in compiled languages, that comes from the fact that parsed one, i.e., where every time it run, it had to read the symbols, call functions to execute them, etc., exactly as written, it is faster to do the math, than to do “if this is true, then add this to ‘value'”. For all I know, compilers recognize that there is math involved and nothing else, and remix the result, into the same thing.
For anyone interested in what I mean, the two constructs look like:
if a > b then value += c;
vs.
value += c * (a > b);
Depending on the language, adjustments may need to be made to the (a > b) part. Some languages will return -1, and some 1, so the above would become ‘value += c;’, in one language, and ‘value -= c;’ in another. I.e., it would switch it from addition to subtraction, so its usually safer to take the “absolute” value, which is “how far is this from 0, without the sign”. A few languages won’t allow the above, since its trying to mix binary with what ever you defined ‘value’ as, in which case you have to use some other thing to get the correct result (which is annoying as hell).
In something like the old Basic language, the math itself takes like 3-4 cycles, but the if/then adds another 3-4, or something, doubling the time taken, and on an old, old, Apple II, this had a major effect on how fast your program was. It also reduced how much memory the program took up, by several unneeded symbols, and since they had even less memory than the famous 640k that PCs had back then… lol
It sounds to me like you’re in violation of one of the following:
First law of program optimization: Don’t do it.
Second law of program optimization (for experts only): Don’t do it yet.
Unless you’re programming just for yourself, the clock cycles you save is rarely worth the confusion you cause others trying to maintain your code.
I’ve been programming for years (including a little assembler on the Apple II at high school), and that method hadn’t occured to me, although I’ve done hairier things, especially when trying to write absurdly compacted Perl.
Returning to topic: The first civil computer Turing worked on (at Manchester, I can’t remember the computer’s name) had cathode ray tube memory: essentially you have a TV screen with light detectors in front of it. The persistence of the glow of the phosphor provides the memory, and the scanning beam usually just rewrites whatever bit was already there, as indicated by the light detectors (unless that memory location is being rewritten.) A side benefit was you could get a second CRT and set it up to take the same signals as the memory CRT, but it doesn’t have the light detector blocking the front. With this, you have the entire memory of the computer visible to you at all times!
(I got this from a different book on Turing, I haven’t read the one under current discussion. Also, I’ve taken a university course on automata and Turing machines, although that was decades ago.)
I think you’re referring to the EDVAC?
The 32 x 32 bit CRT computer at Manchester that Turing worked on was called the Baby. It was meant to be a preliminary model of a much bigger computer, whose name I couldn’t find in a minute of thumbing through the book. They did some amazing work in just 128 bytes of memory, though.
Also, the CRT stored both data and instructions. The realization that the same memory could be used for both was the heart of the von Neumann architecture, which I don’t think was explicitly mentioned in the book, but the Baby may have been the first computer to actually implement a common, electronically stored program and data memory (as opposed to mechanical storage or using separate stores based on the same technology.) Since the program can write to its own instruction memory (which though incredibly useful, is also almost always a mistake), it is possible to write self-modifying programs and compilers.
This Is probably being overly pedantic, and not to downplay Turing’s work at all, but the initial breaking of Enigma was by the Polish intelligence service, just prior to the war. The Germans upgraded the device, and the Poles couldn’t keep up with their earlier bomby machines, what with being invaded and all, and passed their work and the blueprints for their machines to the British and French. The French got invaded as well, so it was up to the British to carry on the work, and Turing was the man to lead the effort.
The Poles get left out of the story a lot; I just feel we should give them their due for breaking early Enigma first, and for keeping the secret that it had been broken all through the war and beyond.
The book did cover the earlier Polish work, which gave Turing and the others a Bletchley Park a big boost. I hadn’t heard of this before. I don’t know if the book overplays or underplays the Poles’ contribution, or is spot on, but it certainly got mentioned.
The German upgrade to their Enigma machines was just weeks before they invaded Poland, and it was apparently all the Poles to do to get their previous work to somewhere safe before they were overrun. Maybe it would make a good movie, courageous heroes fleeing the advancing Nazis with their precious secrets which would eventually win the war. “Based on a true story!”
There was apparently an early movie about Bletchley Park that ignored Turing completely.
“earlier movie”, i.e. some years ago.
If you haven’t been to Bletchley Park yet, then find an opportunity to go.
It’s been converted into a superb museum, with working replicas of both Colossus and the Bombe (both contain some original parts to boot), with about 20 different Enigma machines, a Lorentz cypher machine (the cypher that Colossus broke), the UK’s national computing machine, multiple exhibits about codebreaking and about working at Bletchley during the war, and – at the moment – an Ian Fleming special exhibition (Fleming served at Bletchley early in the war before becoming PA to C – ie Fleming’s real job was the same as Miss Moneypenny’s in the Bond novels/films).
There is, of course, loads of stuff about Turing there, but also a major Tommy Flowers exhibition (engineer that designed colossus).
GCHQ is also progressively transferring over the public parts of its history. It’s hoped that they bring in some of the Clifford Cocks work – notably his independent (prior) invention of what is publicly known as the RSA algorithm.
There is a Turing machine as the Google doodle today! I’m glad I read the book, as I knew what it was right away.