Archive for the ‘bad science’ Category
I knew Andrew Wakefield was full of shit, in a conflict of interest, practicing crappy lab technique, a hypocrite who thought autism was caused by measles viruses when he was in Britain and mercury when he was in America, someone who dealt with people who were promoting quack treatments like giving your autistic kid chemical castration and washing all the electrolytes out of their bloodstream…but I didn’t know precisely what he was proposing as an alternative to the MMR vaccine.
Nor did he, in fact; he hadn’t characterised the agent he proposed to extract from mice infected with measles, inject into goats, sorry, pregnant goats, extract from the goats’ colostrum, mix with human bone marrow cells, and then inject into other people’s children. His patent doesn’t actually say what the “transfer factor” is supposed to be, and his proposed production process wouldn’t select any specific cell type, protein, or other chemical. In so far as anyone has any idea what “transfer factor” might have been, it looks like the likeliest mechanism of action was the same one that caused a recent drug trial to go horribly wrong.
In fact, one of the main barriers to taking it further would seem to be that it might not have been permitted by the Home Office…because it was unnecessarily dangerous and cruel to the goats.
Read the whole thing – you’ll learn quite a lot about human (and goat) immunology, about the man who got Wakefied interested in autism in the first place (he claimed to have cured patients by injecting them with his own bone marrow , before he got into trouble for helping himself to the opiates) and you’ll be amazed that the Royal Free Hospital let him get this far with a project based on the immunology of the 1940s, at best.
It may have been for the best that Wakefield got sidetracked into being a media oaf, rather than actually getting to the point of injecting randomly selected goat cytokines into children.
One of the things that struck me was that it sounded like the sort of thing Soviet scientists might have got up to at the Stalinist nadir. Lysenko, of course, confined himself to plants. The Randi people reckon the Royal Free were influenced by the possibility he might have brought in research funding; a very Thatcherite Lysenko.
OK, so we took the piss out of the Policy Exchange crowd for seeing reds under the seats on the bendy buses. The group rights agenda. But the interesting thing about the Borisbus is that in a sense, it bears Dean Godson and Andy Gilligan out – design and architecture are, of course, deeply political activities. We shape the things we build, and thereafter they shape us, as Winston Churchill said to RIBA (twice – he believed in making aphorisms earn their corn).
Essentially, the new bus – pics here and here– is a bog standard Wrightbus double decker with some fibreglass styling features, meant to evoke the look of the Routemaster; there’s a funny asymmetric front end, a staircase, and an open platform that isn’t actually open, because it is behind a door which will be locked while the bus is in motion. This stuff is pure ornament – it is utterly without function. Neither is there actually going to be a conductor; the existing revenue protection patrols will occasionally be on board, and that’s it.
Now, the thing about adding a lot of nonfunctional stuff for the sake of style is that it has costs. The Postmodernist architects were fascinated by the way Las Vegas casinos and the like were basically huge industrial sheds, covered with playful flourishes of style, plush carpets, neon signs; but the reason why they could get away with this is that a huge clear-span shed is a pretty efficient solution for housing a business process of some kind, whether it’s a semiconductor fabrication line, a giant distribution warehouse, a brewery, or a giant exercise in legalised fraud controlled by Lucky Luciano. The huge plaster likeness of Nefertiti draped in purple neon canted over the entrance at 27 degrees from the vertical isn’t getting in the way of anything.
But this doesn’t work in a setting of engineering rather than architecture. Changing the internal layout of a bus affects its primary function directly; one of the key limiting factors in the capacity of a bus route is how long it takes to load and unload the bus, which determines how long it waits at each stop and therefore how fast it travels. Making people climb the stairs to get in and out has real performance consequences. As pointed out here, when the rear door is shut, anyone trying to get off the bus will have to push past people getting on to use the middle door.
Also, carrying around a platform and a staircase takes up space that could otherwise be used for…well, that could otherwise be used rather than pissed away on content-free curlicues. As pointed out here, the new bus has fewer seats downstairs than a Routemaster despite being 3 metres longer. I thought we were trying to take up less space on the street and improve the turning circle?
Of course, the reason why giant motorway-side warehouses and casinos can be like they are is that they are usually built in places where land is cheap and there is lots of space…like central London, right?
So what does this tell us about the design politics involved? The first, and obvious, point is that design has consequences. As a result of the whole daft crusade, for years to come, bus users will be putting up with a worse quality of service. Frequencies will be lower, because dwell times will be higher. Alternatively, London will just have to buy more buses to maintain frequency, and fares will go up. Using the buses will be a more exasperating and unpleasant experience than it is now (and that’s saying something). Further, people who for whatever reason find the stairs difficult are going to be punished.
Second, it’s the victory of form over content. It’s not a Routemaster; it certainly hasn’t had the years of kaizen that went into the original design and specifically into the hard engineering of it, the engine and drivetrain and running gear. It doesn’t even look much like one, but the key stylistic tropes are there in order to pretend it does. I’m surprised they didn’t stitch a Lacoste croc on it. And, of course, the costs of this shameless fuckery will endure.
Third, the past must have been better. There is really no reason at all to try to make a modern bus look vaguely Routemasteresque other than kitsch and nostalgia, and it’s no better for being Gill Sans/Keep Calm and Carry On kitsch rather than the Laura Ashley version. You bet there’s going to be a lot of this crap in the next few years. (Fortunately, it also looks like the official aesthetic of David Cameron is going to be achingly unfashionable, like an official aesthetic damn well should be.) But if there is any reason to be nostalgic for Routemasters, it should surely be for the unrivalled engineering record of high reliability; being nostalgic for slower boarding times is like being nostalgic for the good old days of rickets. Come to think of it, Tories do that as well.
In conclusion, this is modern conservatism, implemented in hardware, with your taxes. The obsession with PR, spin, and guff in general? Check. The heel-grinding contempt for the poor? Check. The pride in technical and scientific ignorance? Doublecheck. The low, ugly, spiteful obsession with getting one over on political enemies? (It’s of a piece with behaviour like this.) Check.
Key quotes:
(via Boriswatch): “Never underestimate our masters’ obsession with outward form, as opposed to function and content.” That’s Gilligan, of course.
There’s a point where his risible little village idiot act crosses over into a demonstration of overt contempt for the public, and this is it. I propose to refer to him as Shower Jobby from now on, and I would like to see this elsewhere.
Apple’s internal security team may be scary – and especially the name (Worldwide Loyalty Team). But they are as nothing, in terms of creepiness, to this Microsoft web page, which provides the criteria against which MS employees are assessed for their use of humour and the targets they are given to improve. You will not be able to unread this.
In fact, it’s the kind of thing for which the only valid response is to pretend to take it seriously. Why not print out a copy and carry it around? Score your friends against this fine 4×4 matrix chart!
Via this comment, it turns out that the program is based on the ideas of a 70s cult leader who fell out with the Scientologists in a dispute about intellectual property – how very Microsoft of him – and who reconverted his organisation into the management consulting industry. (I’ve often thought a terrorist group should try that one some day.)
The Wikipedia article on the dispute is very funny – two blind men fighting over a comb doesn’t really do justice to the full absurdity of it, as two cult/hucksters duel over the rights to the kind of ideas that shouldn’t be treated so much as property as like toxic waste, or one of those weird codicils that occasionally force some poor swing-voter to fork out for a new church roof. If they were sane, they’d be fighting to get rid of this stuff; but then they wouldn’t be there.
But the really interesting thing is that Werner Erhard’s ideas have already killed one of the great computer-development groups, Doug Engelbart’s Augment Lab at SRI, which dissolved into a stew of project failure and ego wars under their influence. Here’s the money quote, from What the Dormouse Said:
A woman who Bob Albrecht, the People’s Computer Company guru, had been involved with went through the training and came back transformed into a very un-Zen-like creature. She no longer believed that everything was interconnected, but rather had decided that she wanted it all for herself and would do anything to get it.
There’s a key cultural inflection point right there. And I bet Linus Torvalds doesn’t make sure to check that
Do I ever encourage a near party atmosphere because of my comfort with using humor?
always returns False, or worry about finding his personal brand.
I’ve been reading Bruno Latour’s Aramis, or the Love of Technology, a postmodernist account of the failure of a massive French project to develop a Personal Rapid Transit system. Latour’s book contains chunks of fiction, interviews, historical documents, and authorial comment, broken out by the typography – the experience is more like reading a long blog post containing blockquotes from different sources and snarky comments on them than anything else.
It’s a fascinating exploration of the politics of the project, the nature of projects themselves, and the sources of project failure; running from 1969 to 1987, the scheme went from conceptual paper studies to a major prototype by 1973, and eventually built a large-scale test implementation in the mid-80s, before being suddenly cancelled while an intensive test campaign intended to qualify it for deployment was under way. Latour is primarily interested in how the overall concept and much of the technology stayed the same, although its objectives, planned deployment, and resources changed constantly throughout the project.
He argues that, eventually, the crucial issue was that a project is a fundamentally political concept – it has to recruit the support of people and of interest groups in order to progress, and Aramis was a side-project for nearly everyone involved except for two groups – the engineers working on it, and the French Communist Party. Unfortunately for the first group, the contract for large-scale tests was signed as the last act in office of the Communist transport minister before the party pulled out of the Mitterrand government.
This is of course true; a project needs to create its own tribe and its own culture. However, I’m quite ambivalent about the whole concept; not really about its technical or economic aspects, but rather about the idea of urbanness that was built into its core assumptions. PRT emerged in the 1960s as a technological fix to what its American proponents thought was the steady decline of cities – the big idea was a form of high-capacity public transport that would provide point-to-point service without intermediate stops, in a private environment, rather like a car, but without traffic jams or exhaust fumes or road accidents.
The flip side of this comes up again and again in Latour’s interviews with Matra and RATP executives, regarding their assumptions about the passengers and the user-experience studies that were carried out later in the project. Passengers, apparently, wanted more than anything else to be transported from point to point, “without transfers, without thinking“, without other people. Not that any passengers had actually been asked what they thought at this point. Clearly, the political assumptions built into Aramis from the beginning were that moving around a city was basically unpleasant, and specifically because of the presence of other people. Huge amounts of effort were expended on the contradictory task of building a vehicle and a broader networked system that was both user-controlled and designed to keep the user from engaging with it in any way.
Very significantly, when user studies were actually carried out, the public was notably cool on the idea and found the cabins (patterned, on the inside, on the Renault Espace) unnerving and uncanny – rather than being protected from a sinister and menacing urban jungle, they felt isolated in sealed capsules controlled by automated systems, in which they could still be confronted with strangers. The paranoia and declinism that originally motivated the PRT concept was accurately preserved in its architecture and communicated to its potential passengers.
Of course, if you were to ask me about this on the Northern Line or the 271 bus tomorrow evening, I’d probably be significantly more sympathetic to the idea; it’s much easier to enjoy public transport when it isn’t operating at overload-plus. This was also a criticism of Aramis – the RATP managers found it hard to imagine a system working that didn’t use standing passengers as a buffer for peak demand, which is telling in itself. And the PCF’s interest was presumably in the idea of a communal and high-modernist rival to the car that would also be a major technical boost for French industry.
Another interesting but under-discussed angle is that of failed consilience.
While the most active phase of Aramis development was on, other groups of engineers were solving the problems of routing discrete packets around a dense scalefree network, preventing them from colliding, and providing congestion control, load-balancing, and controllable routing metrics. They were, of course, the IEEE-802 and IETF work groups building the Internet. The engineers down the road at Alcatel working on GSM could probably have told them a thing or two, as well. The analogies between the longest prefix match/shortest path wins logic of BGP and the problems of routing Aramis cars are very close, although one problem that doesn’t come up in internetworking is how to return the empties and make sure there is a sufficient free float of vehicles to maintain the service. (You regularly see small vans redistributing the Velib bikes around Paris in order to deal with just this problem.)
Part of the explanation, and another interesting angle, is that there was clearly a massive culture clash between the Matra defence-electronics managers, the RATP railwaymen, and the software developers subcontracted in to eventually write the routing and speed-control systems. Matra representatives repeatedly mention that there was a need for a revolution in microprocessors, although that is precisely what happened every 18 months throughout the project.
Apparently, a related system is under test around Heathrow Terminal 5, due to go live in “spring 2010”. Anyone taking bets?
Although technically speaking releasing testosterone precursors on the world’s trading floors to try to rally valuations wasn’t criminal as such, the affected traders lobbied to declare it illegal. But governments liked the idea too much, and oxygen masks were forbidden.
(Electronic traders were of course immune, but they knew floor traders would be bullish and that was enough for them to become so, too. Soon there was no need to spend on the chemicals anymore.)
I think it was at last 3GSM I suddenly thought of the possibility of someone inventing a sales drug – a pill that would fill you with energy and induce repetitive behaviour (always be closing, right?), whilst at the same time making you hypersensitive to others’ emotions…but completely indifferent to them, and also completely incurious about what you were doing. The street, or the Street, would probably call it Animal Spirit.
And then I jumped under a tram. No. In fact I thought someone should write the book before someone actually invented it, with whatever nightmarish meat-hook consequences that would have. Then I forgot all about it at wheels-up, until now, when the approach of the next 3GSM brought it slithering to mind.
So, there’s this rumour-surrounded gadget that GIYUS wants people to install on their computers as part of the War on Terror. Obviously, I wondered exactly how it worked; did it analyse the Web sites you visit semantically, so as to target its talking points precisely? Did it use some sort of social recommendation mechanism? Also, I was wondering if there was any way of characterising the network traffic it generated and estimating how many people are using it.
So I did the obvious thing and I actually downloaded it. It’s packaged as a Firefox extension (.xpi); extensions consist of JavaScript files for the application logic and XUL (XML User interface Language) for the look’n’feel, all wrapped up in a ZIP archive. If you don’t have the source of one, all you need to do is pass it through an archive tool and extract all files, and then you can read them in a text editor.
And actually, it’s kind of disappointing; no folksonomy, no textual analysis, not even crude keyword matching. It just grabs an RSS feed from ws.collactive.com, passing in the string “GIYUS”, presumably to ensure it gets the right one, checks if any items in it aren’t already cached, and if so, fires a graphical alert containing the message. It’s basically a e-mail list gussied up in Web2.0 finery, with the feature that it’s marginally less trivial to forward the content to nonsubscribers. It doesn’t even appear to spy on your browsing history.
Of course, there could be some server-side magic involved. You can usually get a rough idea of location from an IP address, and a rough idea is probably best in terms of hit-rate (you’ve a much better chance of getting your geotargeting right for “North London” than “Archway”). And you can draw some conclusions from browser credentials – OS, screen, browser type and version etc. For example, perhaps you’d want to serve the red meat civilian deaths are all a fake stuff to MSIE5/6 users in teh US heartland and the Decent Left stuff to Mac users in North London. So I considered actually installing the extension; but then I realised I didn’t actually want a simulated Melanie Phillips on my sofa any more than I wanted the real thing. However, it’s possible to view the feed on the Web anyway, so I checked.
But they may not even be doing that; I’m on a weird niche ISP, with a linux machine, in North London, and the feed I see at http://ws.giyus.org/points/list is deeply generic.
Surely, though, it’s possible to do better than this? I envisage a sort of Web force multiplier, that would analyse the texts you read as you browse and compute some kind of digest hash, and do the same for every link you send anyone else, stashing the hash of each link in a remote server. As you browse, it compares the hash of the current page with the ones in the DB, and returns a list of possibly appropriate arguments – the strength of this being that they could be data, poetry, code, pictures, video, or indeed anything. We could incorporate some sort of social element, too, to keep a check on quality.
Who here knows about corpus analysis? Most of the academic papers my casual search found gave me that “dog listening to music” feeling. What I need is something like a rather bad crypto hash function – one where two texts with different content would produce non-randomly different hashes. Obviously we’d filter the text with a list of stop words like search engines do, so as to strip out the tehs and ands. We could, for example, use (say) the distribution of words in Wikipedia as a common baseline, and measure how the distribution of significant words in the target texts differs from it.
OK, so yer lie detector. It’s been something of a blogosphere hit. And in the comments, we have Nigel, who appears to know something about acoustic signal processing – in the sense of “makes speech recognition systems for Eurofighters”.
It seems that rather than being a signal at a frequency between 8 and 12Hz, the signal you’re interested in is a signal, of that frequency, modulated onto the main signal. So in fact, you could theoretically detect it through a telephone call. I was wrong.
However, that isn’t what Nemesysco’s patent claims, and they vigorously deny that what they are doing is voice stress analysis. It’s not the pitch of any such signal that is discussed in the patent, either; it’s the change in the numbers of thorns and plateaus.
Our acoustic expert says that this could be a way of measuring the signals required for classical VSA, just not a very good one; and anyway, he argues that VSA itself is useless, even if it was VSA they were promising to conduct. And, of course, they deny that this is their methodology. Further, VSA gives only one measurement, one of vaguely-defined stress – not the nine or so Nemesysco claim to get out of this.
Meanwhile, someone who makes the same spelling mistakes as Amir Liberman does showed up in comments to claim there was more, secret technology involved that they hadn’t actually patented. Interestingly, he showed up from the same network as Nemesysco’s Web site. The same network was also the source of a Wikipedia article which got deleted for advertising, in which Nemesysco claimed that their method uses 129 different measurements and isn’t anything like VSA. No, sir. And there weren’t 129 different metrics in their patent…
OK, so what about those identity cards for (some kinds of) foreign nationals? You’ll recall that the Government promised, back in the spring, to have them out and operational in 300 days. As late as July, there were no actual contracts for the job, but they did actually manage to bring in Thales to start work. So how’s it going?
Well, despite the vast cutback in scope and scale, the decision to base it on crappy existing records, and just to forget about the National ID Register for now (thus obviating the whole point)…it’s already over budget by 29% and it’s sliding right, from March 2009 to August 2010. Cracking; the element of the project they specially rushed forward in order to get something, anything working on time has now slid so badly that it’s caught up with the rest of the project.
Meanwhile, the Home Office is issuing 5,400 fraudulent passports a year, among some 200,000 dodgy docs in circulation. Apparently “automated facial recognition” will solve it; this doesn’t make very much sense, as surely the main problem is people submitting genuine photographs of themselves and falsifying the biographical section of the form.
Further, face recognition systems are poor enough (remember the one in Newham that never actually caught anyone?) at positive identification; checking the face provided against the one on file. The failure rate in the Home Office 2004 trials was about 30 per cent. But the IPS and DVLA seem to think they can rely on it to guarantee that the same person isn’t already registered, and do this by matching faces to a database containing tens of millions of faces, taken under all kinds of different circumstances. What kind of false-positive rate can you expect from that?
In fact, it’s worse; if they’re trying to detect multiple applications or applications under false names, the evidence of an honest application will be the absence of a match. So the most common failure mode will result in the document being issued anyway, and there is no way to detect this. And you won’t be able to assume that a match is proof of fraud either, because of the inevitable false positives; so the chance of successfully getting a passport or driving licence in someone else’s name might actually be better.
Recent Comments
World of Harrowell
BORISFEED- An error has occurred; the feed is probably down. Try again later.
Alternate Seat of TYR 