Archive for May, 2003

Greedy Web gets lazy too

Saturday, May 17th, 2003

This idea might be going somewhere — I don’t know. But Danny O’Brien, at least, wrote back as if he thought it had some merit:

Pay me $50, and I’ll do it. Seriously: it’s a great idea, and while similiar things have been tried in the past (notable Brian Behlendorf’s SourceXchange and some others), I think whittling it down to its basics and being a very minimalist middle man might mean that it’ll work this time. Also the fact that a) everyone’s desperate for money these days and b) realised that they’re never going to code up everything they want in life, means that it’s got a better chance now than before.


Saturday, May 17th, 2003

President Bush, of course, is not a junior reporter for the New York Times. So maybe it doesn’t matter if he makes up stories and puts them in the newspaper. After Ronald Reagan, it’s almost a presidential tradition.

Michael Kinsley in Slate . It’s like watching the Statue of Liberty break out in a can-can. Encore! Encore!.

Global Fame

Saturday, May 17th, 2003

In case anyone has wandered over from the Guardian‘s site, I thought I’d put on something to reward them: a quote from Ann Richards, the former governor of Texas, on the crippling effect of oil price rises on the state: “the price of gasoline has gotten so high that Texas women who want to run over their husbands have to carpool”

(from Molly Ivins )

The reason there might by a few fresh readers this morning, is that my column on the Guardian’s web site finally got going yesterday. Since I spent all yesterday afternoon at the funeral of a close friend’s sister, watching his mother faint from grief, over and over, I’m clean out of jokes of my own.

The public misunderstanding of science

Thursday, May 15th, 2003

There’s a survey out from the University of Michiagan, showing that the American public actually understands less about genes and gene therapy than it did in 1990. Actually, what the headline says, nicely illustrating its own point, is that the “Public knows no more about genetics than in 1990”. None of the questions are actually about genetics per se.

The other thing to notice is that the percentage of answers got right has actually diminished since 1990. It’s not that the public understands no more: by any measure, it gets a far smaller proportion of the answers right: in the 1990 survey, a little over a thousand people scored an average mean of 2.7/5 right — more or less what the macaques would do with their keyboards. In 2000, a sample of 1834 people got 1.9/5 questions right.

So ten years of science journalism, and of unprecedented hype about the human genome, conducted by some of the smartest people on the planet, have left the American public worse informed than it was before. I doubt the British case is much better. Life? Don’t talk to me about life.

The greedy web

Thursday, May 15th, 2003

Ben Hammersley invented the lazy web, a way of asking people if they’ve already solved your problem. It’s flashy, and seems to work for a small subset of highly connected geeks but it’s never going to solve the real problem of software for individuals, which is that there’s no market. There’s only monopoly or free.

Within large companies, this doesn’t matter. They can always build their own in house solutions, and that, obviously, is what most programmers are employed to do. Increasing amounts of that stuff is getting built with open source software, or using one of the Scandinavian licences (Qt, mySQL ) which say, ‘this program is free unless you’re trying to make money from it, in which case we want money too’.

But these are toolkits, or means to an end. Most small businessmen wouldn’t have the faintest idea what to do with them. The fairly ambitious open source projects which produce stuff in theory usable by anyone tend also to have ferociously anti-commercial ideologies. There’s no simple way for the small preferences of thousands of users to add up to a medium-sized incentive for one competent programmer to tweak a piece of “free” software in the way that’s supposed to be so easy. Yet there are hundreds of thousands of more or less disgruntled users out there, and thousands of competent programmers.

I was put in mind of this by a student in Australia offering to put up $50 for a decent word count in OpenOffice. There must be thousands of people who’d put up $5 or $10 for that: the web page where I put up a bad word count macro gets at least 100 hits a day.

So what we need is a way to put them in touch with each other: what we need is a greedy web, where people can post requests for little hacks to Open source software, and what they’d be willing to pay; alternatively, if they see a request already made for a feature they’d like, they could vote for it by offering some more money. It’s a sort of auction for something which doesn’t exist until it’s put on sale. Of course, you would need mechanisms to prevent exploitation: there would have to be escrow payments, deadlines, public commitments, and some way of resolving disputes if the software doesn’t perform as expected. But if a hundred people are putting up $5.00, and they get a word count that is slightly inaccurate, they won’t feel half as ripped off and resentful as one man might who had put up $500. And since the money is not paid over until the software is delivered, there’s no scope for scam artistry by the authors.

Any software or hacks produced under this scheme would of course have to be freely available under the licence of the original progam. It would be a nice touch if the commissioners got their names in the comments.

Of course, I couldn’t begin to build such a thing. But it might be incredibly useful. It might even pay for itself if one charged an upkeep fee off the top of the transactions. So I suppose I am appealing to the lazy web to build the greedy web.

cognitive feudalism

Monday, May 12th, 2003

I heven’t yet reread, with laptop to hand, Pascal Boyer, though I know that I will. It is one of those books that requires innocent (at least not guilty) contemplation; but the idea that has stuck with me as I go along is is the sheer disconnection and autonomy of our reasoning processes. This is he uses as his general explanaiton of the religious imagination: our tendency thio think in autonomous, unconscious threads, or deamons, each running simple programs continually.

This needn’t be cognitive dissonance — two threads may not conflict at all, and the dissonance only comes when two incompatible conclusions are generated by two processes (probably both unconscious) acting on the same input. But it does account for the extraordinary ways in which people are fragmentary and inconsistent, without any central authority. This is a particular sort of society of the mind, on which the great baronies are self-sustaining and largely independent, though all will pay homage to the central court. In fact, what I suffer from is cognitive feudalism.

Monday Morning

Monday, May 12th, 2003

News over the weekend: someone in Paignton has actually supplied six monkeys with typewriters, to see how much and of which authors they’d produce. they got bored — the scientists that is — and stopped after six weeks, but not before the macaques had reached a state that takes professionals a lifetime. A spokesman explained: “The monkeys aren’t reducible to a random process. They get bored and they shit on the keyboard rather than type.”

trusted serial killers

Monday, May 12th, 2003

Well, a different war this time. The Guardian’s coverage of the Stakeknife affair has given me the best belly laugh in weeks. However, if one of Gerry Adams’ trusted friends is indeed proved to be not only a tout but a serial killer, the paper writes, discussing the consequences of the discovery. The shock! the horror! That Gerry should turn out to be friends with a serial killer! Next thing, we’ll be told that Martin McGuinness was in the IRA.

When I first went to Belfast, in the early Eighties, I was told by British army sources that Martin “Babes” McGuinness had personally murdered eight people that they knew of, and Danny Morrison, three. Now we know where they got their figures from.

I think the whole story is wonderful news: running the head of the enemies’ murder squad while pretending to go along with their pretence to negotiate in good faith is a better way to deal with terrorists than the alternative, which we also tried, which is simply to shoot, lock up, and torture as many of the bastards as you possibly can. Not morally better, perhaps, but not worse either, and more effective.

Stupid IE tricks

Saturday, May 10th, 2003

This will only work if you are using IE, but it is only fun if you’re not.

If you want to know how it works, or what it does, press “more”.


Error Message

Wednesday, May 7th, 2003

This just appeared in my inbox:

Dear Andrew Brown,

A friend has just sent me your review of Nature via Nurture, in the April 20th Mail on Sunday. I am the daughter B F Skinner put in a box, but it was certainly NOT a larger version of the ‘Skinner Box’ for studying animal behaviour, as you wrote, and your comment makes my father sound heartless and stupid. The baby box – or air crib, as he called it – was simply an enclosed cot, with temperature control, great visibility through glass windows and lots of space to move around. I only stayed in it when I was sleeping (or sometimes playing, as there was lots of room for toys) and it was in no way an experiment. I was a very happy and healthy baby and had a normal upbringing, in fact, with lots of nurturing.

My father never denied the importance of nature, either. He happened to think nurture was more important than most other people of his time. But apart from all that, he was a terrific father whom I loved dearly, and it hurts me when he’s put down in so facile a manner.

Yours sincerely, Deborah Skinner Buzan

I see I must apologise.