10 November 2005

Prof Dawkins and ID cards.

This is a quote from a letter Dawkins wrote to his 10 year old daughter explaining the importance of evidence in making a judgement about whether something is correct or not.

"[there is] evidence, which is a good reason for believing something , and [I] warn you against three bad reasons for believing anything. They are called "tradition," "authority," and "revelation."

It got me thinking about how opponents use the unquestionable importance of 'civil liberties' and 'privacy' without thinking that they have to provide evidence to back up exactly WHY this is important. They are using it as a sort of authority that cannot be challenged. Some however do at least realise there can be a conflict between civil liberties and social justice.

"it's important to decide how much privacy to trade for how much of whatever `social good' the surveillance is supposed to get you."

This is from our friend Chris Lightfoot, from an excellent article in support of speed cameras.

He also says the following;

"While the speed cameras policy has good intentions, it has not been well-handled; and the reaction to speed cameras -- both the visceral response of drivers who like to drive too fast, and that of others who believe that the policy is flawed -- suggests that other, more useful, safety measures may not be welcomed either. Which is sad."

This pretty much is my thinking about ID cards, an excellent policy that has been very poorly handled by the government. I believe it is selfish of a minority to deny the 'social good' unless they can demonstrate there would be significant detriment to their lives. I've yet to be persuaded.

34 comments:

  1. Neil, you're not treating the intellects of participants in this discussion with due respect: you repeated claims which had already been rebutted, without mentioning this. I refer to your claims that biometrics can be somehow scrambled such that they're not a danger to protected witnesses, and that ID cards would help with CNP fraud.

    You're also not playing too fair with this Richard Dawkins stuff, though it's encouraging to see you citing authority against your practice of citing authority rather than evidence.

    You're reversing the burden of proof. You're the one who wants change, so you have to prove why the (net) costs of the ID cards scheme should be borne, or that the scheme will have a net beneficial effect. The 1939-1952 scheme was a (net) loss to the taxpayer, and I've seen no evidence that this new scheme will be otherwise.

    ReplyDelete
  2. "It got me thinking about how opponents use the unquestionable importance of 'civil liberties' and 'privacy' without thinking that they have to provide evidence to back up exactly WHY this is important. They are using it as a sort of authority that cannot be challenged. Some however do at least realise there can be a conflict between civil liberties and social justice."

    Wrong - as Martin points out, you are trying to inverse the burden of proof. The requirement for evidence rests with proponents who wish to change the relationship between the state and the citizen. There is evidence aplenty - but as it doesn't say what you would like to hear, you choose to ignore it.

    As for selfishness, my grandfather sacrificed his health and much of his life for the freedoms this country enjoys - freedoms you would cast aside. If he could face being torpedoed on the Murmansk run, I can face resisting this perfidious bill.

    Your ability to twist an argument out of all recognition would make Alistair Campbell proud.

    ReplyDelete
  3. If I'm not mistaken (and having talked to Chris a considerable amount during writing my pieces referenced by his), the conclusion is that there is no compelling evidence that speed cameras make any change to the accident rate. It is clear however that the speed camera partnerships lie through their teeth with their statistics, as does the ABD & Safe Speed (although in my experience recently Safe Speed has improved, possibly as a result of having it's previous efforts picked to pieces).

    Whilst the original intention 'cut accidents by using speed cameras' was good, they failed to account for pissing off every motorist in the country, and the fact that cameras might not cut accidents. They ignored the first and failed to test or monitor the second - the figures produced by the partnerships have a systematic bias, the size of which can't be estimated because they do not collect the data necessary to measure it.

    More importantly, the speed camera partnerships still do not collect the data required to figure out if cameras work - one can only theorise that the reason for this is someone might discover that they don't[*] and shut down the camera partnerships.


    [*] This isn't a statement of my belief that cameras don't work. It just hasn't yet been shown that they do, so we must still consider the possibility that they don't.

    ReplyDelete
  4. So you're arguing from the (dubious) authority of Dawkins to say that .. we shouldn't accept arguments from authority. That's the funniest thing you've said so far :-)

    ReplyDelete
  5. You might call that an article "in support of speed cameras", but I wouldn't! Indeed, I wrote,

    "But why should motorists trust the government to Do The Right Thing with a system with such potential for abuse when we have already seen them plunging onwards with speed cameras, biometric 'security', and other doomed technology projects without any firm justification for their use?"

    and in a later piece I show that there is no evidence for speed cameras having had a positive or negative effect on the total road fatality rate.

    Perhaps you'd like to correct your article?

    ReplyDelete
  6. Neil,

    you seem to have misquoted Chris here. I must also say that I objected to your decontexualising of my remarks about Sweden in one of the first posts in our exchange. I hope this isn't a habit of yours, and were good could you remedy these sorts of problems.

    ReplyDelete
  7. ...three bad reasons for believing anything. They are called "tradition," "authority," and "revelation."

    The tradition that governments generally work for good. The authority of groups like the police - who will almost always favour the draconian as it helps them avoid the diffcult work they are supposed to be doing. The revelation that converts to biometrics seem to suffer in spite of plenty of evidence that the technology doesn't work reliably.

    As to being selfish - you and your cohorts are selfishly trying to impose an expensive and ill-concieved scheme with no net benefits on me.

    ReplyDelete
  8. Martin; you can't look at the financial cost of this scheme in the short term. Of course the wartime scheme was a financial loss over 13 years, but spread the start-up cost over say 20 or 30 or 40 years and these costs are vastly diminished. Do you think railways and roads would have been built if they had looked at the costs over a 12 year period?

    It also would be difficult at first to quantify all the benefits, but in principle people would have known there were plenty to be had. Without government support, the railways might never have been built or certainly not to the same scale, yet they are of enormous financial benefit to our economy.

    Besides of course, there are not just financial benefits to the scheme. It is difficult to measure all the non-financial benefits from crime reduction for example.

    To be fair to a few of you on here including Martin and Chris, you have not just cited civil liberties as some sacrasanct authority that is unquestionable. You have 'tried' to back it up with evidence. A lot of your supporters don't even try and that was the point I was making by citing Dawkins.

    ReplyDelete
  9. Chris you have duly pointed out that I have misinterpreted your support for speed cameras. But reading the article it certainly seemed you were in support. There is a lot in there criticising opponents of speed cameras.

    You also mention that you support the congestion charge in London. So my main point about you seeing the need to curtail some privacy for the social good stands.

    ReplyDelete
  10. urko: "The tradition that governments generally work for good.

    The authority of groups like the police - who will almost always favour the draconian as it helps them avoid the diffcult work they are supposed to be doing.

    The revelation that converts to biometrics seem to suffer in spite of plenty of evidence that the technology doesn't work reliably."

    It is not a tradition that govts generally work for good. Look at social and economic progress increasing over time and there is your evidence.

    ID cards will not be run by the police, so this authority point is irrelevant.

    Look, I'm under no illusions that biometric technology doesn't need to improve, but it doesn't take a genious to realise how quickly technology has advanced and is still advancing in this area. If the technology is not ready, then the ID scheme will not happen, it will be delayed. You are trying to claim it will NEVER be ready and that is not supported by evidence.

    ReplyDelete
  11. Of course the wartime scheme was a financial loss over 13 years, but spread the start-up cost over say 20 or 30 or 40 years and these costs are vastly diminished.

    It was loss-making on its year-on-year expenses too -- a loss of (I think -- I don't have my full notes here) about £500,000 per year on running costs of £700,000 per year. (Source is a 1950 document titled "The Future of National Registration" -- I'll get you the full reference if you're interested.)

    There is a lot in there criticising opponents of speed cameras.

    Criticising the opponents of something is not the same as supporting it. It is perfectly possible -- as with most of the debate on speed cameras -- for everybody involved in the public debate to be wrong.

    ReplyDelete
  12. So you think we should get rid of speed cameras then?

    ReplyDelete
  13. "The Future of National Registration" -- I'll get you the full reference if you're interested.)"

    Chris, if you could find us a full reference, I would appreciate it. Of course, a scheme with a photo or other biometric security would be of much less financial value.

    ReplyDelete
  14. Neil,

    you continue to insult me and Chris.

    Why do you use scare quotes around the word "tried" when saying that we tried to adduce evidence?

    And no, Chris and I don't play up the civil liberties aspects. So let's stop talking about that until you've met our objections. If I were to use your tactics, I'd take all the worst arguments of the stupidest, nastiest, maddest, most ignorant and dangerous people and say that you were wrong because people you agree with on one issue (the principle of ID cards) are wrong on another (bankrupting our pension scheme by keeping the darkies out). This type of argument is called ad hominem tu quoque, and it's bollocks, and you know it is bollocks, and you keep doing it, so stop.

    As should be obvious from my statements elsewhere, I regard human rights as a dangerous ideology rather than some sacrosanct totem. It really is insulting to read that I'm being lumped in with these people. Why do you do it?

    I'm not interested in the arguments about the principle of ID cards. I don't believe in a Platonic ideal of ID cards. I don't believe in whatever it is people are talking about when they say "identity". I am only interested in the practical arguments about the government's actual schemes. I'm not interested in schemes the Government has already ruled out. I'm just interested in the actual, practical, concrete, real-world costs and benefits of the schemes that might actually be introduced.

    We've done ID cards in principle. Can we talk about ID cards in the real world now? You have a lot of unanswered points to deal with, and the burden of persuasion rests with you.

    ReplyDelete
  15. Of course, a scheme with a photo or other biometric security would be of much less financial value.

    That's what we've been saying! I think you left out a negative there :-)

    ID cards will not be run by the police, so this authority point is irrelevant.

    While strictly that's true - it'll be run by the Home Office which is told what to do by the police, as we saw recently. The police will also have privileged access to it, and eventually the power to demand ID cards. I notice that one of the examples cited by http://www.identitycards.gov.uk/ is the criminal records bureau.

    I've been reading their stuff recently, which is interesting. They too seem to have fallen for the "there is this (new) regulatory requirement to check identity/CRB/etc, the cost of this would be reduced by ID cards" line, without thinking holistically about the regulation.

    ReplyDelete
  16. By the way, you've never addressed my point about solving credit fraud without ID cards which I've posted twice now.

    ReplyDelete
  17. Martin, Chris, I am not insulting either of you, and I'm sorry if you've got that impression. I specifically mentioned that the argument didnt apply to ALL opponents including you and Chris.

    But you have to admit there are a fair nunber on your side that do believe in this abstract notion of civil liberties without feeling they have to justify why or what exactly they are on about.

    Just as Chris has mentioned that a lot of racists support ID cards, isn't he trying to tar my side of the argument by association? I never said he was, but using your logic, he must be.

    I too am only interested in the practical arguments for and against the government's scheme. I am quite happy to concentrate on that, so answer my question about why you lump start-up costs of a long term scheme in with the short term running costs.

    ReplyDelete
  18. Peter Clay: "Of course, a scheme 'with' a photo or other biometric security would be of much less financial value"

    Obviously I meant to put 'without'.

    "By the way, you've never addressed my point about solving credit fraud without ID cards which I've posted twice now."

    I have answered it on another thread somewhere. But I'll answer it again.

    People should only be able to enrol on cardholder-not-present schemes face to face using their ID cards.

    ReplyDelete
  19. "I have answered it on another thread somewhere. But I'll answer it again.

    People should only be able to enrol on cardholder-not-present schemes face to face using their ID cards."

    That's not an answer, that's begging the question! I proposed a means of solving fraud (that is, make it the banks' problem and let the private sector figure it out, and protect third parties from being wrongly pursued) without ID cards, and you've not said anything about it, you've just used it as an excuse to demand ID cards.

    I have this nasty feeling from the Home Office stuff that ID cards are a solution in search of a problem. There hasn't been a holistic consideration of the problems they are alleged to solve.

    Do you really think there is no other way to solve the problem than mandatory biometric ID cards?

    ReplyDelete
  20. "Do you really think there is no other way to solve the problem than mandatory biometric ID cards?"

    I think you are right, there are other ways of tackling it, but ID cards would tackle it as well, and they have other benefits. I wouldn't be against a combination of strategies.

    ReplyDelete
  21. What is this "enrol" on CNP schemes? CNP is a characteristic of some transactions. The two I know from personal experience are ordering stuff online with my credit card, and to let you run a tab at a bar. (They want to be able to bill your card if you walk off without paying, which is fair enough).

    Now you've been told about these before, so why are you pretending that "enrolment" is necessary, desirable or possible?

    ReplyDelete
  22. It should be possible for banks to bar CNP transactions to anyone who hasn't enrolled with an ID card. I don't see why that would be difficult. That way people get a choice to opt out of CNP transactions if they choose to. This would make it even more difficult for fraudsters.

    ReplyDelete
  23. That doesn't help. The card holders who have gone through the additional hurdle on account creation are just as vulnerable to CNP fraud as they were before.

    .. and we've already established that, by definition, we can't do an ID check against this new information since the cardholder isn't there.

    ReplyDelete
  24. Neil,

    you don't seem to understand; CNP is not a type of fraud. It is a type of transaction. It is a desirable type of transaction. Without CNP, you can't efficiently do things which people want (like ecommerce and bar tabs).

    Restricting its use to people who have gone through one more ID check at some time and place other than those involved in the transaction is a net loss, and has no impact on CNP fraud.

    ReplyDelete
  25. At the moment fraudsters prefer face to face fraud because it is quicker, easier and they are less likely to be caught. (no worries about collect addresses etc.)

    When ID cards come in, face to face fraud will be much more difficult, so fraudsters are likely to switch their attention to CNP fraud. Because CNP transactions will be far more risky, people should be able to 'opt out' of having that type of transaction available to them unless they have enrolled for it with an ID card.

    You are quite right that this doesn't directly affect those who have enrolled, but at least they know about the extra risks they are taking on.

    It is likely a lot of people will opt out of CNP transactions, so any card a fraudster steals or copies, he won't know whether it is opted in or out of CNP, until he tries to use it.

    This presents an extra risk of being caught for the criminal, as if suspicions are raised, the address could be checked before they even get any goods. In this way it will deter CNP fraud as well as face to face fraud.

    ReplyDelete
  26. Because CNP transactions will be far more risky, people should be able to 'opt out' of having that type of transaction available to them unless they have enrolled for it with an ID card.

    It's been pointed out to you that the notion of "enrolling" for CNP transactions is meaningless. Why do you repeat it?


    Where you say that you expect CNP transactions to become riskier -- I assume you mean relative risk here? -- who do you expect will bear this risk -- cardholder, vendor or issuing bank? Who should it be? What do you think the costs of fraud to the non-risk-bearing parties will be? And what should they be?

    ReplyDelete
  27. "It's been pointed out to you that the notion of "enrolling" for CNP transactions is meaningless. Why do you repeat it?"

    I can't see why a bank couldn't let people 'opt out' of the CNP type of transactions if they want to. I don't see how this would be difficult.

    "who do you expect will bear this risk -- cardholder, vendor or issuing bank?"

    This argument about who directly bears the costs of fraud is factitious.

    Whether it is the bank, govt or cardholder. It is the consumer that pays in the end in added prices.

    ReplyDelete
  28. By "factitious", do you mean "fictitious"?

    Do you understand what is meant by the "variance" of a statistical distribution?

    ReplyDelete
  29. No, I did mean factitious.

    "Do you understand what is meant by the "variance" of a statistical distribution?"

    I understand what is meant by it, what is it's relevance in this discussion?

    ReplyDelete
  30. I believe it is selfish of a minority to deny the 'social good' unless they can demonstrate there would be significant detriment to their lives. I've yet to be persuaded.
    There is no social good so it doesn't need to be denied - there would be no point. As to detriment - it will cost all of us lots of money that could be much better spent on targeted cime prevention and detection measures aimed at the real problem areas

    ReplyDelete
  31. OK. Your argument is that all the costs of fraud will, ultimately, be born by cardholders. That's a respectable position. Let's suppose that the total cost of fraud is £x in a given year, and there are N cardholders. What should be the distribution of costs born by each cardholder? In particular, what should be its mean and its variance?

    ReplyDelete
  32. Of course Chris, not even the cardholder bears the total costs. It is the consumer who pays higher prices when the vendor bank charges increase. The point I make is that ANY reduction in £x reduces this total cost to the consumer. Since higher prices is a more regressive penalty than higher taxes, it is more egalitarian for the government to bear the cost.

    ReplyDelete
  33. Can you answer the question, please?

    ReplyDelete
  34. Well I don't know what the present distribution is so I can't give a definite answer to that, apart from to say that it should move towards as egalitarian a distribution as possible.

    ReplyDelete