Sunday, April 25, 2010
French bloggers, Scud missiles, etc
There's a new strategy blog about, this time a French one. They have an interesting discussion about the suggestion/rumour/story that Hezbollah might be trying to acquire Scud missiles. They're dubious about it, although open to the suggestion that the organisation might be developing its own inter-service politics, with the big rocket people perhaps constituting the "air force". Relatedly, this pair of Tom Ricks posts is stuffed with interesting links about the 2006 war. In the light of those, you might almost think that any effort to acquire something like a Scud, with its huge transporter-erector-launcher truck, would be more of a deception or disinformation tactic than anything else.
project failure
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 Latour's book contains chunks of fiction, interviews, historical documents, and authorial comment, broken out by the typography - the experience is more like reading a long blog post containing blockquotes from different sources and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?
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?
self-compulsion notice: diplomats
I have been scraping things with Scraperwiki this weekend. Which made me wonder about this post on Spyblog about that Israeli diplomat who one of the Milibands told to pack his bags. Mark quotes from the official London Diplomatic List, which turns out to be issued every month as a PDF by the Foreign Office protocol bureau, and which lists all embassies in London and their members. You can see where this is going, and it's going to involve pdf2xml...
Further adventures in globalisation
So you might remember that Thai demonstrators invaded the brand-new airport there a while ago, establishing a huge Ballardian protest-camp among the glass walls and retail space and soft-xray terrorist detectors. Their movement went on to spray the prime minister's house with their own blood, collected in buckets by their medical wing. Clearly, they have a certain style.
Which made me think when I saw this BBC story; how much science-fiction would you need to get from being stuck at the same airport due to northern Europe getting a fine dusting of Iceland, while the Redshirts and the cops and the No Colour Movement - colour revolutions have clearly reached some sort of logical end point - duke it out downtown, to actually getting the queues involved in the revolution? (The other way round is much easier, and amounts almost to a cliche.)
Which made me think when I saw this BBC story; how much science-fiction would you need to get from being stuck at the same airport due to northern Europe getting a fine dusting of Iceland, while the Redshirts and the cops and the No Colour Movement - colour revolutions have clearly reached some sort of logical end point - duke it out downtown, to actually getting the queues involved in the revolution? (The other way round is much easier, and amounts almost to a cliche.)
thuds and screams in Wapping
So, that election. I should be out delivering leaflets; but my leaflets haven't turned up. Such is life in the sinister Lib-Dem election machine.
More interestingly, it's been a week of truth. We kicked off with David Yelland's and Michael Wolff's pieces about the likely panic in the Murdoch world about the Lib Dem surge and, more broadly, the possibility of their pet candidate, complete with Andy Coulson as personal representative, failing to win. That was interesting, but you could have been forgiven for a certain scepticism.
Then, however, the Murdoch world decided to throw a live demonstration (in more ways than one, thinking about their visit to the Indy). Kicking off, the Tories announced "a new Get Clegg strategy". Rather, it might have been truer to say that they "announced" it; at the time, the news reporting was merely that it had been announced. What was really meant by this became obvious on the morning of the second debate, with the synchronised wave of abuse from the right-wing press.
By lunchtime on the next day, however, they had been called on the issue, by name:
The direct quote is from the Liberal election coordinator, Danny Alexander; the rest is the Guardian's. You have to love the impressively yellow bit about "there is no evidence that Conservative HQ..." after the direct statement that, yes, Gideon personally made the rounds and handed out the talking points.
Because, of course, the Lib Dem surge has at least achieved one thing - it's provided an opportunity to observe the media-political complex working in real time. We can't tell what transpired between Osborne and the pet editors, but we do know that an unexpected third-party surge happened at the beginning of the week, the Tories promised smears, and that certain newspapers all delivered them on the due day. Input-output analysis.
At least we now know, for a fact, that there are newspapers in the UK that accept direct orders from politicians, and we also know which ones. Not that the list is a surprise. It's also interesting that the Murdochs still think the Times has to observe slightly different standards; it sat out the story. Clearly, self-delusion does actually act as a real check on some people's behaviour. The Times kids itself it's still a newspaper, and therefore is somewhat more like one.
Another thing; since Osborne got caught, the Tories have adjusted the fire somewhat. Ken Clarke told the Telegraph that they might consider going into coalition with the Liberals; the Observer, in its Obscurer mode, headlined the same thing over an interview with David Cameron, in which he didn't really say that - but presumably his PR team must have accepted it. The whole thing has had a strangely communist feel; a Power Struggle in the Inner Party, which the people follow through odd snippets of certainly misleading news-like data.
The good news: ignorance is no longer excusable. The even better news is: there's an app for that, or rather a Greasemonkey script.
(More here, and here, among others.
In an environment characterised by uncertainty, the best predictions are often the ones based on humour and caricature; they get past the shared illusions and get right to the irrational core.
More interestingly, it's been a week of truth. We kicked off with David Yelland's and Michael Wolff's pieces about the likely panic in the Murdoch world about the Lib Dem surge and, more broadly, the possibility of their pet candidate, complete with Andy Coulson as personal representative, failing to win. That was interesting, but you could have been forgiven for a certain scepticism.
Then, however, the Murdoch world decided to throw a live demonstration (in more ways than one, thinking about their visit to the Indy). Kicking off, the Tories announced "a new Get Clegg strategy". Rather, it might have been truer to say that they "announced" it; at the time, the news reporting was merely that it had been announced. What was really meant by this became obvious on the morning of the second debate, with the synchronised wave of abuse from the right-wing press.
By lunchtime on the next day, however, they had been called on the issue, by name:
"George Osborne needs to come clean as to whether he himself was personally responsible for this negative media smear campaign, which is now backfiring spectacularly with voters."....
Osborne met some political editors on Monday and discussed the party's response to the Lib Dem surge.
In one paper, a strategist was reported as expressing the hope that the media would do the Tory party's dirty work. There is no evidence that Osborne made this remark or that Conservative headquarters fed any story to any paper.
The direct quote is from the Liberal election coordinator, Danny Alexander; the rest is the Guardian's. You have to love the impressively yellow bit about "there is no evidence that Conservative HQ..." after the direct statement that, yes, Gideon personally made the rounds and handed out the talking points.
Because, of course, the Lib Dem surge has at least achieved one thing - it's provided an opportunity to observe the media-political complex working in real time. We can't tell what transpired between Osborne and the pet editors, but we do know that an unexpected third-party surge happened at the beginning of the week, the Tories promised smears, and that certain newspapers all delivered them on the due day. Input-output analysis.
At least we now know, for a fact, that there are newspapers in the UK that accept direct orders from politicians, and we also know which ones. Not that the list is a surprise. It's also interesting that the Murdochs still think the Times has to observe slightly different standards; it sat out the story. Clearly, self-delusion does actually act as a real check on some people's behaviour. The Times kids itself it's still a newspaper, and therefore is somewhat more like one.
Another thing; since Osborne got caught, the Tories have adjusted the fire somewhat. Ken Clarke told the Telegraph that they might consider going into coalition with the Liberals; the Observer, in its Obscurer mode, headlined the same thing over an interview with David Cameron, in which he didn't really say that - but presumably his PR team must have accepted it. The whole thing has had a strangely communist feel; a Power Struggle in the Inner Party, which the people follow through odd snippets of certainly misleading news-like data.
The good news: ignorance is no longer excusable. The even better news is: there's an app for that, or rather a Greasemonkey script.
(More here, and here, among others.In an environment characterised by uncertainty, the best predictions are often the ones based on humour and caricature; they get past the shared illusions and get right to the irrational core.
Saturday, April 10, 2010
oh dear
The Tories' iPhone app ('cos they're modern and stuff) works by sending a plaintext e-mail to callafriend@myconservatives.com, it turns out.
bubbles
Everyone's reviewing Dean Baker's False Profits. I contribute.
I strongly recommend the book, agreeing with D^2 that it's important that he names the guilty men, reminds you that they're guilty, and keeps on naming them. (In fact, if anything, he drowns the fish - if you weren't sure if there had been a housing bubble, you'll be sick of hearing about it.)
But I think there's an important point here; so long as there is capitalism, entrepreneurs will every so often become over-extended, chase the price of something up a mountain, and then panic and crash. In fact, perhaps even outside capitalism; aren't the investment disasters of the Soviet Union something similar? Usually, however, it's possible to look at this process philosophically; as J. K. Galbraith said, nothing but money is being lost, and the people who are losing it usually have plenty. Residential property, though, is special - it alone brings the full destructive power of a leverage-fuelled financial crisis within reach of the common man.
As the Despair poster has it, none of us is as stupid as all of us together. A housing bubble's effects are therefore very widespread, and individually catastrophic to a degree that isn't true of financial crises restricted to bankers, investors, and entrepreneurs. They also have a special feature, which is that they interact with the tools of macro policy closely. Once the bubble passes a certain point, a significant chunk of the buyers are no longer able to cope with the normal range of fluctuation in the interest rate, and disaster is certain. This is essentially Minsky stage 2 - when capital gains become a necessity in order to service debt.
Once this stage is reached, for many of the people involved, it doesn't matter whether the final crash comes sooner or later - they're going to be ruined. For society as a whole, it would probably be better if it came sooner, but there are obvious reasons why this is unlikely to happen.
Keynes once said, in Economic Perspectives for our Grandchildren, that he hoped economists would become "humble competent people like dentists". Daniel Davies said not so long ago that economics might, if it was lucky, eventually become a branch of control systems engineering. An important concept in both is stability, but stability is often misunderstood. It doesn't mean the absence of change, but rather, the ability to recover to a normal state. James Lovelock's insight about ecosystems was arguably that they are stable - they eventually recover - but that there was no guarantee that recovery would be a good thing.
In a housing bubble scenario, eventually the berserk run-up in prices does indeed go into reverse - but it does so through widespread bankruptcy, unemployment, and systemic bank failure. This is the same whether the bubble bursts on its own, or the bubble is deliberately burst. The recovery mode of the system is indistinguishable from the worst-case scenario. As the purpose of a system is what it does, the fact we get into these scrapes is telling in itself.
Here, Felix Salmon excerpts a range of fascinating charts from the US National Housing Survey. A critical detail; even people who are currently behind on their mortgages or even in the process of repossession believe, by large majorities, that property is a good investment and that it is less risky than Treasury bills. Clearly, it will need more than just the biggest bubble in economic history, leading to the worst economic crisis since the Great Depression, to convince anyone that there is a problem. (There's also an excellent chart making the point that the market is still dear historically by about 20%.)
An interesting counter-example is here; it seems that specific regulation against some of the worst bubble-generating practices can help a lot. And I think that changing the failure modes is probably a good way of changing the system. Dean Baker answers reviewers here; he also points out that although Spain's banks didn't blow up, the housing bubble there was sufficient in itself to flatten the economy.
I strongly recommend the book, agreeing with D^2 that it's important that he names the guilty men, reminds you that they're guilty, and keeps on naming them. (In fact, if anything, he drowns the fish - if you weren't sure if there had been a housing bubble, you'll be sick of hearing about it.)
But I think there's an important point here; so long as there is capitalism, entrepreneurs will every so often become over-extended, chase the price of something up a mountain, and then panic and crash. In fact, perhaps even outside capitalism; aren't the investment disasters of the Soviet Union something similar? Usually, however, it's possible to look at this process philosophically; as J. K. Galbraith said, nothing but money is being lost, and the people who are losing it usually have plenty. Residential property, though, is special - it alone brings the full destructive power of a leverage-fuelled financial crisis within reach of the common man.
As the Despair poster has it, none of us is as stupid as all of us together. A housing bubble's effects are therefore very widespread, and individually catastrophic to a degree that isn't true of financial crises restricted to bankers, investors, and entrepreneurs. They also have a special feature, which is that they interact with the tools of macro policy closely. Once the bubble passes a certain point, a significant chunk of the buyers are no longer able to cope with the normal range of fluctuation in the interest rate, and disaster is certain. This is essentially Minsky stage 2 - when capital gains become a necessity in order to service debt.
Once this stage is reached, for many of the people involved, it doesn't matter whether the final crash comes sooner or later - they're going to be ruined. For society as a whole, it would probably be better if it came sooner, but there are obvious reasons why this is unlikely to happen.
Keynes once said, in Economic Perspectives for our Grandchildren, that he hoped economists would become "humble competent people like dentists". Daniel Davies said not so long ago that economics might, if it was lucky, eventually become a branch of control systems engineering. An important concept in both is stability, but stability is often misunderstood. It doesn't mean the absence of change, but rather, the ability to recover to a normal state. James Lovelock's insight about ecosystems was arguably that they are stable - they eventually recover - but that there was no guarantee that recovery would be a good thing.
In a housing bubble scenario, eventually the berserk run-up in prices does indeed go into reverse - but it does so through widespread bankruptcy, unemployment, and systemic bank failure. This is the same whether the bubble bursts on its own, or the bubble is deliberately burst. The recovery mode of the system is indistinguishable from the worst-case scenario. As the purpose of a system is what it does, the fact we get into these scrapes is telling in itself.
Here, Felix Salmon excerpts a range of fascinating charts from the US National Housing Survey. A critical detail; even people who are currently behind on their mortgages or even in the process of repossession believe, by large majorities, that property is a good investment and that it is less risky than Treasury bills. Clearly, it will need more than just the biggest bubble in economic history, leading to the worst economic crisis since the Great Depression, to convince anyone that there is a problem. (There's also an excellent chart making the point that the market is still dear historically by about 20%.)
An interesting counter-example is here; it seems that specific regulation against some of the worst bubble-generating practices can help a lot. And I think that changing the failure modes is probably a good way of changing the system. Dean Baker answers reviewers here; he also points out that although Spain's banks didn't blow up, the housing bubble there was sufficient in itself to flatten the economy.
Kursk
Kursk was a bit of a disappointment. A submarine control room as the setting of a play isn't a bad idea - the movies worked that out many years ago - and putting it on as promenade theatre through the simulated sub is a cracking one. But, not quite.
It did remind me to check the (excellent) Wikipedia article on the loss of the Kursk, which answered my question. The problem is that the story doesn't really provide for a good drama from the viewpoint of a British submarine; even if you accept they were present, had they decided to surface at once and steam up to the Pyotr Veliky, the best thing that could have happened would have been to launch the ineffectual Russian rescue attempt a few hours earlier, which would have changed nothing. Most of the dead were dead within seconds; the survivors survived for days, almost long enough for the eventual British and Norwegian rescue effort to save them.
This leaves the story as a pure sea-piece; the isolation of the submarine, the role of the captain, the character conflicts, navy culture, the details of control-room procedure. In fact, the set's two-level structure, laid out around the central search periscope, isn't all that far off the Navy's original submarine simulator in design. In the original, the mockup control room was on the lower level, with the periscope rising through the ceiling into a room where the images required for the training scenarios were projected onto the walls.
You could make a case for secrecy being the main theme, but again, it doesn't quite work. A minor note is that there's a fair bit of Americo-scepticism about; the presence of two Los Angeles-class boats in the area is pointedly briefed as the American "threat".
Seen as science-fiction, though, it holds up better. An SF writer, whose name I forget, once said that there weren't any wars in his books because the universe was enemy enough.
It did remind me to check the (excellent) Wikipedia article on the loss of the Kursk, which answered my question. The problem is that the story doesn't really provide for a good drama from the viewpoint of a British submarine; even if you accept they were present, had they decided to surface at once and steam up to the Pyotr Veliky, the best thing that could have happened would have been to launch the ineffectual Russian rescue attempt a few hours earlier, which would have changed nothing. Most of the dead were dead within seconds; the survivors survived for days, almost long enough for the eventual British and Norwegian rescue effort to save them.
This leaves the story as a pure sea-piece; the isolation of the submarine, the role of the captain, the character conflicts, navy culture, the details of control-room procedure. In fact, the set's two-level structure, laid out around the central search periscope, isn't all that far off the Navy's original submarine simulator in design. In the original, the mockup control room was on the lower level, with the periscope rising through the ceiling into a room where the images required for the training scenarios were projected onto the walls.
You could make a case for secrecy being the main theme, but again, it doesn't quite work. A minor note is that there's a fair bit of Americo-scepticism about; the presence of two Los Angeles-class boats in the area is pointedly briefed as the American "threat".
Seen as science-fiction, though, it holds up better. An SF writer, whose name I forget, once said that there weren't any wars in his books because the universe was enemy enough.
better than it is
An interesting isotope is detected in the CRU report fall-out plume. Apart from the very high concentrations of concern-troll, tone-troll, and pure drivel, there is something worth learning from.
There are hardly any scientists today who don't do some programming of some sort; there's not much science that doesn't involve churning through really big data sets. As a result, there's a lot of it about. Which reminds me of this Eric Sink post from 2006, about the distinctions between "me-ware, us-ware, and them-ware". Me-ware is software that you write and only you use; us-ware is software that is used by the same organisation that produces it; them-ware is software that is produced by a software company or open-source project for the general public.
There's a gradient of difficulty; the further from you the end-user is, the less you know about their needs. On the other hand, if you're just trying to twiddle the chunks to fit through the ChunkCo Chunkstrainer without needing to buy a ChunkCo Hyperchunk, well, although you know just how big they are, you're unlikely to spend time building a pretty user interface or doing code reviews. Which only matters up to a point; nobody else would bother solving your problem.
But this can bite you on the arse, which is what happened to the climate researchers. It's fair to say that if you're processing a scientific data set, what actually matters is the data, or the mathematical operation you want to do to it. You won't get the paper into Nature because you hacked up a really elegant list comp or whatever; they won't refuse it because the code is ugly. Anyone who wants to replicate your results will probably roll their own.
This is OK, but the failure mode is when the political equivalent of Brian Coat comes snooping around your #comments or lack of them. Perhaps I should tidy up the Vfeed scripts while I'm at it.
For this reason, many software professionals encountering science software for the first time may be horrified. How, they ask, can we rely on this crude software, developed in primitive conditions - by amateurs, working with such poor tools and such poor understanding of the field? This is a common reaction to GISTEMP, and is exactly the reaction which many critics have had, some very publicly, to the software published with the CRU emails. Such critics do have a point. Science software should be better than it is. Scientists should be provided with more training, and more support. But consider the uses to which science software is put. Most
software written by scientists:
* consists of tiny programs;
* which will only ever be run a small number of times;
* over the course of a few weeks as it is being developed;
* by the scientist who wrote it;
* on data gathered by that scientist's team;
* concerning a scientific field in which that scientist is expert;
* to perform data processing on which that scientist is expert; and will be discarded, never to be used again, as soon as the paper containing the results is accepted for publication.
There are hardly any scientists today who don't do some programming of some sort; there's not much science that doesn't involve churning through really big data sets. As a result, there's a lot of it about. Which reminds me of this Eric Sink post from 2006, about the distinctions between "me-ware, us-ware, and them-ware". Me-ware is software that you write and only you use; us-ware is software that is used by the same organisation that produces it; them-ware is software that is produced by a software company or open-source project for the general public.
There's a gradient of difficulty; the further from you the end-user is, the less you know about their needs. On the other hand, if you're just trying to twiddle the chunks to fit through the ChunkCo Chunkstrainer without needing to buy a ChunkCo Hyperchunk, well, although you know just how big they are, you're unlikely to spend time building a pretty user interface or doing code reviews. Which only matters up to a point; nobody else would bother solving your problem.
But this can bite you on the arse, which is what happened to the climate researchers. It's fair to say that if you're processing a scientific data set, what actually matters is the data, or the mathematical operation you want to do to it. You won't get the paper into Nature because you hacked up a really elegant list comp or whatever; they won't refuse it because the code is ugly. Anyone who wants to replicate your results will probably roll their own.
This is OK, but the failure mode is when the political equivalent of Brian Coat comes snooping around your #comments or lack of them. Perhaps I should tidy up the Vfeed scripts while I'm at it.
actual Iranian music news
I was needling Spencer Ackerman about this but didn't get a rise. It's applicable to Noel Maurer too.
No-One Knows About Persian Cats is a cracking little film; it's a pseudo-documentary about Iran's music underground, by the Kurdish director Bahman Ghobadi and a small who's who of Iranian music. As a result, it could almost have been designed for Spackerman in the way Jeremy Clarkson said Vulcan 607 could have been designed for him.
One thing that comes through are the permanently-operating factors in the human terrain. For example, there's always a fixer - the guy who doesn't actually contribute any music themselves, but does know people who know people who have access to studio time and hall bookings and dodgy government permits. It's the Tony Wilson ethic. Hamed Behdad plays him as someone of permanent charm and near perfect unreliability, never clear whether he's totally committed to success or on the point of making off with the funds - one reading of the grim ending is that he's the grass.
The metal band's guitarist works - like Tony Iommi - in a metalworking factory, and the band rehearse in a shed full of cowshit on the edge of town, although paradoxically their lyrics are all about positive thinking. The rappers are slightly thuggish and given to lyrics like "the class struggle oppresses us!" which may have worked better in the original. The indie band are a bit painfully sensitive and notably more middle-class, the sort of people these guys are thinking of.
So far, so good; anything that reminds us that Iran is not actually Nazi Germany or the far side of the moon is politically welcome. So much of this is immediately recognisable if you've ever sat in a Mini with rust holes and a 1x12 Valvestate box on your lap, with a curry balanced on the top.
Of course, making music in an authoritarian society has its special problems. Everyone except the rappers is desperate to leave and the plot revolves around rounding up passports, visas, and means of payment, as the East Germans used to say. And getting the Ministry of Virtue permit to actually put on a gig. In the meantime, there's a constant round of rehearsals in cellars and in breezeblock sheds on rooftops; incredibly complex informal building seems to be a bit of a feature of Tehran.
And there's a sticky end at the hands of the militia, or not quite at their hands enough to prove it. In that way the police tend to have.
The BBC has the soundtrack as streams here.
No-One Knows About Persian Cats is a cracking little film; it's a pseudo-documentary about Iran's music underground, by the Kurdish director Bahman Ghobadi and a small who's who of Iranian music. As a result, it could almost have been designed for Spackerman in the way Jeremy Clarkson said Vulcan 607 could have been designed for him.
One thing that comes through are the permanently-operating factors in the human terrain. For example, there's always a fixer - the guy who doesn't actually contribute any music themselves, but does know people who know people who have access to studio time and hall bookings and dodgy government permits. It's the Tony Wilson ethic. Hamed Behdad plays him as someone of permanent charm and near perfect unreliability, never clear whether he's totally committed to success or on the point of making off with the funds - one reading of the grim ending is that he's the grass.
The metal band's guitarist works - like Tony Iommi - in a metalworking factory, and the band rehearse in a shed full of cowshit on the edge of town, although paradoxically their lyrics are all about positive thinking. The rappers are slightly thuggish and given to lyrics like "the class struggle oppresses us!" which may have worked better in the original. The indie band are a bit painfully sensitive and notably more middle-class, the sort of people these guys are thinking of.
So far, so good; anything that reminds us that Iran is not actually Nazi Germany or the far side of the moon is politically welcome. So much of this is immediately recognisable if you've ever sat in a Mini with rust holes and a 1x12 Valvestate box on your lap, with a curry balanced on the top.
Of course, making music in an authoritarian society has its special problems. Everyone except the rappers is desperate to leave and the plot revolves around rounding up passports, visas, and means of payment, as the East Germans used to say. And getting the Ministry of Virtue permit to actually put on a gig. In the meantime, there's a constant round of rehearsals in cellars and in breezeblock sheds on rooftops; incredibly complex informal building seems to be a bit of a feature of Tehran.
And there's a sticky end at the hands of the militia, or not quite at their hands enough to prove it. In that way the police tend to have.
The BBC has the soundtrack as streams here.
Friday, April 09, 2010
coffee and stupidity
Deeply unprofitable Crooked Timber thread produces an interesting point. Not so much "why do libertarians hate Fairtrade coffee?", but why on earth do they insist on lying? From the link:
The problem here is paragraph two; I've not attempted to estimate the cross-elasticity of demand for Fairtrade and non-Fairtrade coffee, and it's not the kind of thing I'd try unless someone was paying me to do it. But it's perfectly simple to consult the Fairtrade Labelling Organisation International standardisation documents and see if they actually forbid full-time employment or set a maximum limit on the size of farms. Here's the standard on the use of hired labour. Requirement 1.5.1.13:
So - no. Not even close. 180 degrees out, in fact. The point arises downthread that the great majority of work on a coffee farm is seasonal in nature, so Cowan couldn't possibly have been right even if you let him have his own facts. This is marginally unfair to him; it's not his prose, but rather something he's (very) favourably quoting. The original source is here; note the cites back to Cowan and to people at George Mason University, the Koch Industries school.
For some reason, I still followed a link to his blog after this, to this interesting sociological result. Apparently, if you've been to jail, you're more likely to self-identify as black when you leave jail then when you went in. This is interesting, but then numskull racist twit Steve Sailer showed up in the comments. As far as I can make out from his semi-coherent comments, he's very angry about this, which is odd for someone who believes in biological racism (surely it wouldn't matter? your DNA, after all, doesn't care which census box you tick).
Further, Sailer's freakout reminded me of the old joke about the Wee Free minister who hears that a hippie commune has moved into the next island along. He listens icily as the various novel features are described - the long music, the loud hair, the ruthlessly commercial communists, etc. Eventually they tell him the newcomers believe in free love. "Free love?" he explodes. "That could lead to mixed dancing!" Similarly, it's clear from Sailer's comments that he's opposed to rape in prison because he thinks it might lead to multi-culturalism. And does he think about it. All the time, it seems.
Perhaps his mate Anthony Browne from PolEx could offer him counselling as part of the Big Society?
By guaranteeing a minimum price, Fairtrade also encourages market oversupply, which depresses global commodity prices. This locks Fairtrade farmers into greater Fairtrade dependency and further impoverishes farmers outside the Fairtrade umbrella. Economist Tyler Cowen describes this as the “parallel exploitation coffee sector”.
Coffee farms must not be more than 12 acres in size and they are not allowed to employ any full-time workers. This means that during harvest season migrant workers must be employed on short-term contracts. These rural poor are therefore expressly excluded from the stability of long-term employment by Fairtrade rules.
The problem here is paragraph two; I've not attempted to estimate the cross-elasticity of demand for Fairtrade and non-Fairtrade coffee, and it's not the kind of thing I'd try unless someone was paying me to do it. But it's perfectly simple to consult the Fairtrade Labelling Organisation International standardisation documents and see if they actually forbid full-time employment or set a maximum limit on the size of farms. Here's the standard on the use of hired labour. Requirement 1.5.1.13:
All regular work is undertaken by permanent workers...The objective is that as often as possible, work is undertaken by permanent workers. Only work that is added to usual work levels during peak seasons may be undertaken by seasonal workers.
So - no. Not even close. 180 degrees out, in fact. The point arises downthread that the great majority of work on a coffee farm is seasonal in nature, so Cowan couldn't possibly have been right even if you let him have his own facts. This is marginally unfair to him; it's not his prose, but rather something he's (very) favourably quoting. The original source is here; note the cites back to Cowan and to people at George Mason University, the Koch Industries school.
For some reason, I still followed a link to his blog after this, to this interesting sociological result. Apparently, if you've been to jail, you're more likely to self-identify as black when you leave jail then when you went in. This is interesting, but then numskull racist twit Steve Sailer showed up in the comments. As far as I can make out from his semi-coherent comments, he's very angry about this, which is odd for someone who believes in biological racism (surely it wouldn't matter? your DNA, after all, doesn't care which census box you tick).
Further, Sailer's freakout reminded me of the old joke about the Wee Free minister who hears that a hippie commune has moved into the next island along. He listens icily as the various novel features are described - the long music, the loud hair, the ruthlessly commercial communists, etc. Eventually they tell him the newcomers believe in free love. "Free love?" he explodes. "That could lead to mixed dancing!" Similarly, it's clear from Sailer's comments that he's opposed to rape in prison because he thinks it might lead to multi-culturalism. And does he think about it. All the time, it seems.
Perhaps his mate Anthony Browne from PolEx could offer him counselling as part of the Big Society?
trolls of the Ayatollah
Laura Rozen takes us to meet the Iranian government's loyalist bloggers.
You bet they do. It would be offensive to speculate exactly who we all know would end up doing this job in a hypothetical fascist Britain. I'll leave it up to you, although I will point out that at least two bloggers who would be regulars on Newsnight and in the British Gazette's opinion pages in that scenario...already are regulars in the MSM.
I don't know if they're doing anything technical to favour their bloggers and their trolls over the other guys; if the loyalists are more likely to be hosted in Iran, the policy of slashing international bandwidth while leaving the networks up might help. But that's not what interests me right now.
What role, politically, do trolls play? On one hand, it's clearly possible to use the Internet as a mobilisation tool for good, or at the very least, for nihilistic shit-flinging. Examples; this slightly disappointing interview with the Obama campaign's CTO, and this blitz on a bunch of bigots' facetwitspace accounts, respectively. Or this scientific paper; oddly, when a random Internet person actually did some climate science they didn't find that it was all made up.
But on the other hand, there's a great towering mountain of drivel, a spuming, stinking Eyjafjallajökull of bullshit - an Icelandic or Hawaiian eruption, one that keeps burbling on without working up enough pressure to explode, but does keep belching toxic gas.
Personally, I suspect that the use of Internet pond life in politics is that it's a way of tapping the energies of people who otherwise wouldn't get involved, just as lefties tend to hope it might be the same thing. I just differ on which group of people are being mobilised. It's hard to get The Authoritarians to initiate anything; they're obsessed with leaders by definition. And it's also unlikely that you'll get people who are convinced of the futility of collective action to start a movement. Further, this guy wasn't going to take to the streets, was he?
Get them in front of a keyboard, pass them some talking points, though, and they're happy to bombard selected targets with abuse. I further observe that we have about three major examples of this - one is the US, and specifically the Bush re-election campaign, another would be Russia in the early Putin years (Andrew Wilson's classic Virtual Politics (I reviewed it here) is good on the importance of the temnik talking points system), and the latest would be Iran today.
The hardliners have fielded hundreds of pro-regime bloggers, most writing in Persian, emphasizing particularly the importance of supporting the Supreme Leader. The bloggers range from students to clerics, with many claiming to be members of the Basij and children of war veterans or martyrs. Virtually all are hard-line or extremist in their views. Some bloggers appear to be popular and often draw many comments from their posts.
You bet they do. It would be offensive to speculate exactly who we all know would end up doing this job in a hypothetical fascist Britain. I'll leave it up to you, although I will point out that at least two bloggers who would be regulars on Newsnight and in the British Gazette's opinion pages in that scenario...already are regulars in the MSM.
I don't know if they're doing anything technical to favour their bloggers and their trolls over the other guys; if the loyalists are more likely to be hosted in Iran, the policy of slashing international bandwidth while leaving the networks up might help. But that's not what interests me right now.
What role, politically, do trolls play? On one hand, it's clearly possible to use the Internet as a mobilisation tool for good, or at the very least, for nihilistic shit-flinging. Examples; this slightly disappointing interview with the Obama campaign's CTO, and this blitz on a bunch of bigots' facetwitspace accounts, respectively. Or this scientific paper; oddly, when a random Internet person actually did some climate science they didn't find that it was all made up.
But on the other hand, there's a great towering mountain of drivel, a spuming, stinking Eyjafjallajökull of bullshit - an Icelandic or Hawaiian eruption, one that keeps burbling on without working up enough pressure to explode, but does keep belching toxic gas.
Personally, I suspect that the use of Internet pond life in politics is that it's a way of tapping the energies of people who otherwise wouldn't get involved, just as lefties tend to hope it might be the same thing. I just differ on which group of people are being mobilised. It's hard to get The Authoritarians to initiate anything; they're obsessed with leaders by definition. And it's also unlikely that you'll get people who are convinced of the futility of collective action to start a movement. Further, this guy wasn't going to take to the streets, was he?
Get them in front of a keyboard, pass them some talking points, though, and they're happy to bombard selected targets with abuse. I further observe that we have about three major examples of this - one is the US, and specifically the Bush re-election campaign, another would be Russia in the early Putin years (Andrew Wilson's classic Virtual Politics (I reviewed it here) is good on the importance of the temnik talking points system), and the latest would be Iran today.
export any
So we were talking about China exporting its internal chaos, while also importing Indian internal chaos. Then, the good folk at 31 Jin-rong Street, Beijing gave us a practical example. That would be AS4134, CHINANET-BACKBONE, aka China Telecom's long lines/Internetworking division. Starting at 1558 GMT, they leaked a very large number of other people's routes into the global Internet. According to Martin Brown of Renesys, the incident affected some 31,847 routes in 10 minutes, or several times the number normally announced from 4134.
Anyone who accepted one of the leaked routes would have seen their traffic to that network go to China Telecom first, and only then to their destination, presuming that China Telecom's internal routing was valid. Otherwise it would have gone nowhere.
The key thing to remember here is that Internet routers know where to route your traffic because they tell each other which routes they have. Network A - in this case AS4134 - tells its peers that it can route to network B directly and to C via D or E, but preferably D. The peers pass on this information to their peers. Eventually, this means that any router connected to the Internet can have a full copy of the routing table for the entire Internet at any moment, should it need it. The downside of the Border Gateway Protocol which governs this is, however, that it doesn't perform any verification of the routing updates; like a lot of Internet engineering from the 70s and 80s, it relies on trust. More to the point, it relies on trust in others' competence.
Occasionally, bad things happen, as when Pakistan accidentally routed all inbound traffic to YouTube into the censorship proxy at Pakistan Telecom. In this case, somehow, China Telecom issued 31,847 routing updates to the world at large. An important distinction here is that between internal and external BGP; it's possible that a huge national backbone operator might have offered its customers a seriously large number of routes, indeed a full route-view. But they certainly shouldn't have offered them to the wider Internet - hence the idea of a routing "leak".
We've blogged before that the Chinese Internet isn't characterised by "cyberwar" or even by censorship, but rather by chaos. This is a physical, real-time example of internal chaos within China being exported to the rest of the world. As a result, not just because of BGP issues but also of spam, etc, quite a lot of networks filter all or most netblocks inside China; here's a sample access control list.
Anyone who accepted one of the leaked routes would have seen their traffic to that network go to China Telecom first, and only then to their destination, presuming that China Telecom's internal routing was valid. Otherwise it would have gone nowhere.
The key thing to remember here is that Internet routers know where to route your traffic because they tell each other which routes they have. Network A - in this case AS4134 - tells its peers that it can route to network B directly and to C via D or E, but preferably D. The peers pass on this information to their peers. Eventually, this means that any router connected to the Internet can have a full copy of the routing table for the entire Internet at any moment, should it need it. The downside of the Border Gateway Protocol which governs this is, however, that it doesn't perform any verification of the routing updates; like a lot of Internet engineering from the 70s and 80s, it relies on trust. More to the point, it relies on trust in others' competence.
Occasionally, bad things happen, as when Pakistan accidentally routed all inbound traffic to YouTube into the censorship proxy at Pakistan Telecom. In this case, somehow, China Telecom issued 31,847 routing updates to the world at large. An important distinction here is that between internal and external BGP; it's possible that a huge national backbone operator might have offered its customers a seriously large number of routes, indeed a full route-view. But they certainly shouldn't have offered them to the wider Internet - hence the idea of a routing "leak".
We've blogged before that the Chinese Internet isn't characterised by "cyberwar" or even by censorship, but rather by chaos. This is a physical, real-time example of internal chaos within China being exported to the rest of the world. As a result, not just because of BGP issues but also of spam, etc, quite a lot of networks filter all or most netblocks inside China; here's a sample access control list.
Wednesday, April 07, 2010
the voices of randomness
My Markov chain blog occasionally brings forth wonders.
It's incredible how the paranoia makes itself felt. A Lovecraft theme here, of course.
Martin "Asian supermen are our rightful masters - submit!" Jacques, meanwhile:
The a speech of David Cameron would be facing in this country who hope that the Jews generally and poisonous Jewish millionaires the Centre for many Americans are long tentacles of the little platoons of the gathering darkness.
It's incredible how the paranoia makes itself felt. A Lovecraft theme here, of course.
Martin "Asian supermen are our rightful masters - submit!" Jacques, meanwhile:
The Chinese believe that a trade war drawing in the country's centre of power relate to a long drawn out process of the Uighurs: duties and equally fine wine a likely possibility.
Indeed a characteristic of government is running scared.
exporting Chinese chaos, importing Indian chaos
The new Shadowserver Foundation report is out; everyone has ooh'd over folk stealing the Dalai Lama's e-mail, etc. Others have pointed to the concentration on the Indian military establishment.
Technically, all that's interesting here is that the attackers used mass market Internet services, like Yahoo! Mail, as transports for their botnet command-and-control messages, and that they made a lot of use of dodgy PDF files as an attack vector, usually personalised to the target. Also, their network visualisations suggest that registering DNS names with minimal consequences is very important to the system as a whole - each name is requested by many, many bots and is used to identify many, many Web servers. Fast-flux would seem to be a crucial node in the system.
Politically, the first major insight is that it's India that's the top priority, and India beyond the Tibetan exile institutions. India, excluding the Tibetan targets, vastly dominates all categories of everything in the report. Indian institutions are the top targets - as well as a wide range of defence organisations, the Indian Railways are in there. There's no word on Bollywood yet. So far, so clear - rather than worrying about much fancied dissidents or Tibetans, the strategic priority is the local peer competitor.
On page 33, however, detailing what was extracted from the Indian National Security Council Secretariat:
Now, it's very possible that this is just the effect of dragging a big scoop through a flock of top Indian Civil Service presentations. On the other hand, why would China prize data on Shokly's lot? Various options are possible. One, they're splittists causing irresponsible destabilisation and obstructing construction - and they want to either know all about them in case of war, or else know all about them because they're both a bunch of modern thinkers very keen on big coal mines.
Another would be that the Chinese side is interested in the Naxalites as a potential model for their own internal chaos, as Jamie Kenny would say. China's forced-draft modernisation - as Samuel Huntington might have said in one of his periodic bouts of moral self-abasement - is constantly reacting with a tough working class that expresses its protests by burning down Communist Party offices. Like India, they have deeply uneven development, and a strange relationship between top-down, authoritarian modernisation and federalism. Perhaps they're wondering what form a super-MGI would take?
Yet a further option would be that the Ghostnet team are interested in the Naxalites because they're rebels themselves. They may be working for the Party, but it's very clear from the report that the line between the commercial spammer/botherd business and the spooks is blurred. The same techniques and the same people and the same places appear in both. In one way, this suggests that the crimeware world is a significant resource for the Party and the government; in another way, it suggests that they're beholden for this to an unruly and uncontrollable gang of botnet masters who'd frame their grandmothers as paedophiles for money, and do worse for bragging rights.
Technically, all that's interesting here is that the attackers used mass market Internet services, like Yahoo! Mail, as transports for their botnet command-and-control messages, and that they made a lot of use of dodgy PDF files as an attack vector, usually personalised to the target. Also, their network visualisations suggest that registering DNS names with minimal consequences is very important to the system as a whole - each name is requested by many, many bots and is used to identify many, many Web servers. Fast-flux would seem to be a crucial node in the system.
Politically, the first major insight is that it's India that's the top priority, and India beyond the Tibetan exile institutions. India, excluding the Tibetan targets, vastly dominates all categories of everything in the report. Indian institutions are the top targets - as well as a wide range of defence organisations, the Indian Railways are in there. There's no word on Bollywood yet. So far, so clear - rather than worrying about much fancied dissidents or Tibetans, the strategic priority is the local peer competitor.
On page 33, however, detailing what was extracted from the Indian National Security Council Secretariat:
In addition to documents containing the personal and financial information of what appears to be the compromised individual, the exfiltrated documents focus on India's security situation in the states of Assam, Manipur, Nagaland, and Tripura, as well as the Naxalites, Maoists, and what is referred to as "left-wing extremism".
Now, it's very possible that this is just the effect of dragging a big scoop through a flock of top Indian Civil Service presentations. On the other hand, why would China prize data on Shokly's lot? Various options are possible. One, they're splittists causing irresponsible destabilisation and obstructing construction - and they want to either know all about them in case of war, or else know all about them because they're both a bunch of modern thinkers very keen on big coal mines.
Another would be that the Chinese side is interested in the Naxalites as a potential model for their own internal chaos, as Jamie Kenny would say. China's forced-draft modernisation - as Samuel Huntington might have said in one of his periodic bouts of moral self-abasement - is constantly reacting with a tough working class that expresses its protests by burning down Communist Party offices. Like India, they have deeply uneven development, and a strange relationship between top-down, authoritarian modernisation and federalism. Perhaps they're wondering what form a super-MGI would take?
Yet a further option would be that the Ghostnet team are interested in the Naxalites because they're rebels themselves. They may be working for the Party, but it's very clear from the report that the line between the commercial spammer/botherd business and the spooks is blurred. The same techniques and the same people and the same places appear in both. In one way, this suggests that the crimeware world is a significant resource for the Party and the government; in another way, it suggests that they're beholden for this to an unruly and uncontrollable gang of botnet masters who'd frame their grandmothers as paedophiles for money, and do worse for bragging rights.
Subscribe to:
Posts (Atom)
