Interview: John Graham-Cumming

Interviews

The ethics behind a pardon for Alan Turing have proved highly divisive. Andrew Gregory talks to John Graham-Cumming about making Gordon Brown apologise.

Open source developer and writer John Graham-Cumming was able, through a Downing Street online petition, to persuade the Gordon Brown Labour-led government to issue an unequivocal apology for the gross indecency conviction of Alan Turning in 1952. After admitting a sexual relationship with a man, Turing was unable to continue work as a code breaker at GCHQ, as his security clearance was withdrawn. Two year later he killed himself. Linux Format caught up with the Graham-Cumming to discuss open source development, debugging and why he opposes the move for a pardon, which has been given via royal decree.

Linux Format: Well done for the apology. How did that come about?

John Graham-Cumming: I saw a mention of the fact that it was Alan Turing’s 90-somethingth anniversary birthday. I’d always known about the story of Turing because I’d studied computer science, and I’d also done cryptography stuff, so I knew quite a lot about him. It’s one of these stories that always made me so angry; the way his life ended, the loss of him. I just wrote a rant on my blog: “Britain should apologise, blah blah blah,” and somebody emailed me, some random person, and said: “You should set up a petition on the Number 10 website”. I didn’t know that that existed, and the reason I didn’t know is that I had just moved back to Britain after having lived abroad for 15 years, so I didn’t know there were petitions….

So I went on there and thought: “Alright, I’ll do that”, thinking that only 500 people would sign it or something. And, initially, only about 500 people did sign it; because you had to be in the UK, you had to care enough about the story, so… and then I just worked on media, got the story out and it just grew and grew and grew, until the BBC wrote about it. And then it just exploded, because the BBC has such credibility.

LXF: What’s the threshold at which the government has to make some sort of response? Something like 100,000 signatures?

JGC: I can’t remember whether there was a threshold necessary in the previous system. But we never have got there, because it got to about 32–33,000 signatures, and then I literally got an email, saying: “Can you please call Number 10 Downing Street?”, which I thought was a hoax. So the first thing I did was I Googled the phone number, because it could have been some random joker, and it was actually the switchboard at number 10. So I called, and got through to an assistant to the Prime Minister, who said: “We’re doing the apology tonight. We just want to read you the text, make sure you’re OK with it, and then Gordon wants to talk to you. She read me the text, and I was fine with it; it apologised to Alan Turing, and of course to the other people who were convicted, that was the important thing.

About 10 minutes later, my mobile rang and it was Gordon Brown. I had him on the phone, and I’d not been very well, I’d had the flu and I’d been in bed, so I was feeling a bit groggy. Then I had the Prime Minister on the phone, so I was really kind of starstruck. So I wasn’t very talkative. And he’s not a very talkative guy, he’s kind of an awkward person in fact. And the first thing he said to me, literally the first thing he said, was: “Hello John. I think you know why I’m calling you.” And I’m really glad that I did. Can you imagine: it’s like when the teacher at school says [adopts portentous teacher tone] “You know what you’ve done John, you know what you’ve done. I don’t have to tell you what you’ve done.” So we had this very very difficult conversation, just because the two of us were not quite sure how to talk to people on the phone. It was very amusing.

LXF: He does look awkward on the telly, but I’ve heard he’s awesome in the flesh.

JGC: I actually thought when I was talking to him on the phone that because he was not the sort of polished politician – you know what I mean, Tony Blair would have been very good at that sort of thing – that you could imagine going out and having a beer with him just like that. It would have been much easier to talk to him down the pub, we could just have hung out for a bit. So that was the story of that. So there was the apology, and they put it out on 10 September, and the next day it was news.

John Graham-Cumming being interviewed by Andrew, lovely.  

LXF: Have you had any involvement in the pardon (which is a very different thing).

JGC: I’ve been asked to be involved, and I’ve said that I won’t be involved, because I actually don’t think it’s a good idea. I’ve opposed it without making a huge fuss of opposing it, and the reason I oppose it is that the term ‘pardon’ seems to imply that wrong was done by Turing, and I don’t think that any wrong was done by Turing. Even if legally that’s not quite what the pardon means – because the pardon forgives us, the pardon removes any sentence – well that doesn’t help him anyway. The apology itself was symbolic, so if you’re looking for a symbolic gesture, it’s there. And I thought that the government, the current government, introduced this bill called the Protections of Freedom Act, which is now law, which causes any gay man who was convicted to have their conviction basically erased if they ask.

If you are alive today and you have that criminal conviction you just write to the Home Office and say: “I want you to discard it” and they just do. That protects the living. People who are living under that law, who were convicted [of gross indecency], that’s dealt with that side of it, and I just don’t think that the pardon adds anything new.

Now, obviously, there are some people who do think that it is a big symbolic gesture, and the government now seems to be supporting it, having previously been against the idea of it.

LXF: They announced something last week (the private member’s bill tabled in the Lords had its third reading, in the House of Commons, on October 23, 2013).

JGC: Yes. In a sort of very parliamentary way they said that they’re not going to oppose anything and that if you don’t introduce any amendments then the bill will go to the floor [for a vote]. They won’t stop it becoming something that people can vote on, so maybe there will be a pardon. I don’t think it’s the right thing to do.

You won’t find anybody from Turing’s family supporting the idea of a pardon if you look in the press, and I think that’s significant. When you look at other pardons, for example the shot at dawn guys from the First World War, the families of those people were actively campaigning for their family members, and I think that makes a tremendous difference.

One of the things that’s annoying in the current pardon campaign is that, particularly I think Lord Sharkey, who’s spearheaded it, has said it’s the pardon that Turing’s family deserves, and they’re not actually asking for one. It all feels a bit political and agenda-y. But, you know. I feel like the 2009 thing that I did was very important, and so be it. I think it’s the wrong gesture, it’s the wrong way to go about it. And I also think it’s bad that they’re not asking for a pardon for all the other men who were convicted.

LXF: No pardon for Oscar Wilde then, to pluck one name out of the hat.

JGC: It’s very interesting you should say Oscar Wilde, because one of the things that to me is distasteful about the current pardon campaign is that there was a letter from Stephen Hawking and a number of other eminent scientists in the ///Daily Telegraph// arguing that Turing should be pardoned because of his exceptional ability. It actually says that – it says it in Latin, but it says it. And I really find that offensive, because what you don’t want to do is undo something that was a terrible inequality by creating a new inequality, which would be “if you’re the right sort of gay, it’s OK – we’ll get you off”. I just find that to be completely the wrong message: “It’s OK to be gay if you’re a genius; if you’re not, well, sorry.” I don’t like that side of it.

LXF: It does seem rather unfair on the gay people who were alive when there were no Nazis to fight.

JGC: So you can imagine Oscar Wilde getting a pardon for being a brilliant writer and satirist, but what about the guy who’s just an ordinary bloke? He doesn’t get let off because he didn’t do something amazing. I find the whole thing a bit distasteful, but that’s that.

 

LXF: Let’s move on to the peak of technological development.

JGC: Return of the Jedi is that? [Graham has a wild theory that the high-water mark of technological advancement coincides with the release of the third installment of the ///Star Wars/// trilogy. NB: correlation does not imply causation].

 

LXF: I find it really hard to believe, and I know that’s the gist of what you’re talking about, that it is so unbelievable.

JGC: The gist of my talk is not that there’s nothing been invented since 1983, because of course there have been all sorts of amazing things done since then. But the real gist of it is that it’s very easy for people in the computer industry to not look backwards at what’s been done before, and think they’re coming up with something brand new.

I think the computer world is very good at forgetting the past. If you talk about the mainframe era, it’s almost pejorative, right? Mainframes are these slow, useless machines. And if you actually go and look at what IBM did in System 360, and what companies did, there was a huge amount of innovation in terms of languages, in terms of design, much of which has repeated itself and gone forward. So I think there’s a lot you can learn by looking at the past, and I also think that you can save yourself repeating the mistakes that have been made before. Our field is fairly small; it’s quite plausible that you can learn most stuff about how computers work, from the electrons upwards.

 

LXF: We ran a tutorial a few issues ago about designing a RAM circuit. Jonathan Roberts, our writer, spent ages learning the basics, and he drew these diagrams of switches, showing how to store something like four bits of memory in RAM. Scale it up enormously and you have the 4 or 8GB of RAM from a modern system.

JGC: It turns out that it’s not all that complicated at some conceptual level. We tend to marvel at our devices: “Wow, I’ve got an iPhone with so much memory in it!”, and, OK, it is amazing that we’ve managed to miniaturise all this clever stuff, but on the other hand, a lot of what that machine is doing is essentially just a sped-up version of something that was done many, many years ago.

I don’t want to depress people, it’s more to say that there’s a great foundation that you can build on, and do think about the past, because it’s interesting. But my main point is that the only way in which we have completely failed to do anything is in debugging. Our missions are just unreliable, and we haven’t really got better at it. Software crashes and we just deal with it. I have apps on my phone that randomly crash.

This is an area in which there are great strides to be made. You really could help programmers. I just hope they do it before I get much older, so I don’t have to keep debugging software.

 

LXF: There’s a strong monetary incentive to building a better debugger. It must have the potential to save God knows how many man-hours.

JGC: Can you imagine what the overall economic effect of more reliable software would be? It would be gigantic! Even if you could reduce by a small percentage the number of times that software crashes, I bet you could calculate that the effect on the economy would be huge.

I think debugging stuff is really hard – it’s a lot easier to make Facebook, or whatever. It takes a special kind of person to want to improve debugging, or to develop a new way of understanding. For example, concurrent systems: why they crash is rocket science.

 

LXF: Do you think the fact that so much has been done before in the field of computing… if that were more widely understood, do you think that it would have implications for the patent jungle?

JGC: I think that there are some implications, but more deeply, the real problem with software patents is the obviousness. In patents, you have this idea that something has to be non-obvious.

The term that patent lawyers use is “someone ordinarily skilled in the art”. They mean someone who has experience in that area. For something to be patentable, it has to be non-obvious to that imaginary person. So when you think about how much computing repeats itself, and someone who was ordinarily skilled in the art, a lot of things would be obvious that are patented today, as if they were not obvious.

There’s a famous example that I’d have to check up on, which has to do with a mouse pointer on a screen and what to do with what’s underneath it. If you’re moving a mouse pointer around, there’s something underneath it. There’s a patent for copying what’s underneath it into a temporary buffer somewhere, and copying it back when the mouse moves a bit; and there’s another one for exclusive OR the mouse pointer with whatever’s underneath. How many ways are there to do that anyway? How is it patentable, that sort of thing? If you look at things like Engelbart’s talk, where he’s clicking on stuff, he at some point embeds a document inside another document…

 

LXF That’s the Mother Of All Demonstrations?

JGC: Right. It is an hour and 40 minutes long and it’s quite slow, because he’s presenting and talking and trying to explain stuff that no-one outside of SRI had seen. Right at the beginning of the talk, he says: “I do all my work on this computer”, which is a completely radical idea, right? Then he shows himself making a shopping list on screen. At the time, it was completely amazing that he was doing this. And he saves it, and adds a numbered list, and labels. And then he adds links between things, and he draws a map of all the shops he has to go to, and he links from the shop to the bit of the list that he has to buy from that shop. And, also, he’s in San Francisco doing all this, talking to a machine in Menlo Park, so it’s all being done over a network.

It’s a little bit slow to start with, because we’re so used to being able to do all those things, but if you take a step back and think about it, it’s amazing that all this everyday stuff was being done as far back as 1968.

So, he’s got a mouse – which he invented so he could have a way to point – he’s got a keyboard with him. He’s also got this weird chording keyboard that’s got three buttons on it, which he uses for some other operations. It’s amazing. There’s also this other demonstration that I came across from 1967, where there’s a woman who uses a light pen to highlight some text, grabs it and moves it down to the bottom.

 

LXF: Hang on, she’s clicking on a link on-screen?

JCG: She actually wants to copy and paste something, and she does it just by swiping the pen across the screen. What? This is when I was born!

But the deeper point was, a lot of this stuff has been done before, and particularly in some of the really hard areas, where processor architecture, IBM did so much of that work. Actually, even Charles Babbage’s design had a pipeline in the analytical engine. So while the CPU, which he called the mill, was doing its work, the rest of it was actually fetching the next set of data on to the bus, which is not actually a bus, it’s a stack of cogs. So he designed data pipelining because he could speed the machine up by half a cycle. That was 1841 or something.

I’ve been programming for quite a while, and you see these things come back. One thing I forgot to mention in the talk was the Go programming language. This is interesting because Google very specifically talks about the heritage of that language and says “this is based on CSP, from 1978”.

And I think, actually, that there’s sort of a lesson in there, which is that rather than them claiming: “This is all new and shiny”, they’re actually saying “We’ve chosen the best things… we’ve looked at what we think is stuff that works, and we’re going to bring it together into this new language”.

 

LXF: We were speaking to Chris DiBona about this yesterday, and he said that the two things he’s most proud of are Go and Android.

JGC: Go is wonderful. We use it a lot at work.

 

LXF: Do you reckon that that sort of recycling of ideas is helped by free software?

JGC: I think it helps a lot. If you go back and look at a lot of the stuff that was invented in the 50s and 60s, particularly, you’ve either got things like Xerox or SRI, where they had research labs, right, where there was a tremendous amount of collaboration going on between different fields: Ethernet was invented at Xerox; the stuff that Doug Engelbart did was at SRI. Those were in some sense open source environments, because they were academic environments essentially. They were within the company, but they were sharing.

IBM was a bit of an exception because it had tremendous power in terms of the number of people it had, it could do a lot of things internally, but the reality of things is that my point about the age of great productivity is that open source really made that take off. It’s not just ideas, but it’s the code that implements the ideas and starts to work with it. Open source is of tremendous importance.

You should follow us on Identi.ca or Twitter


Your comments

Randy

<a href="http://all-money.biz">http://all-money.biz</a> Нет такого дня <a href="http://mebel-v-smolenske.ru">http://mebel-v-smolenske.ru</a> чтобы за ним не наступил вечер. <a href="http://smol-imperia.ru">http://smol-imperia.ru</a>

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Post new comment

CAPTCHA
We can't accept links (unless you obfuscate them). You also need to negotiate the following CAPTCHA...

Username:   Password:
Create Account | About TuxRadar