Saturday, June 21, 2008

Chuck and Bruce

No, this is not a post about the new marriage thing in Kali. There's apparently some speculation on who would win a fight between Bruce Schneier and Chuck Norris.

Yes, Chuck can divide by 0 by using a roundhouse kick, while Bruce is known to have counted to infinity -- several times! -- in search of prime numbers greater than infinity (of which he knows several). Still, the outcome of single combat between them seems obvious:

As the two titans square off, Lemmy apparates into the battlefield, reduces them to squirming masses of flesh by yelling at them to get back to using their powers for the good of mankind, then smokes a pack of Marlboros through a bong filled with Jack Daniels, drinks the bong, makes sweet love to three or four groupies, and goes back on tour.

Ch. Duh!

Labels:

|

Tuesday, June 17, 2008

Heh

|

Saturday, June 07, 2008

Transparency, again

In this post, I introduced two variations on a definition of transparency:
  • How are decisions arrived at? Who has the decision authority? What is the basis of a decision? How may the decision be appealed?
  • Authority and responsibility must lie at the same locus.
But depending on the context, transparency also means a few other things. With respect to commercial activities, transparency means that you get what
you pay for and it is known to you what you are paying for it. Seen from this standpoint, surprisingly, one of the more transparent transactions in which you will ever engage is the purchase of gasoline. They put the price on a great big sign out front; the price even includes all federal and state taxes; and you control the trade at every step. You could argue that it is less than transparent since the amounts explicitly going to the state and feds are not usually spelled out. But since everyone in the local area has the same burden built in, what difference does it make where it goes? It's the cost to you that matters.

The opposite of this might be the purchase of a house: first you negotiate with the previous owner, then you negotiate with the bank, then you have to pay title insurance, mortgage insurance, transaction fees, flood insurance, and possibly a few other things before you take possession. Later, you find that there are tax advantages and disadvantages (depending on where you live). Other complex deals are comparable: the purchase of a cell phone with a plan and a car with financing involve bundling, hidden costs, fees, taxes, and so on.

For the most part, though, purchasing stuff in a modern economy has become so much more transparent than it was here in the past, or the rest of the world even now. Most of my daily transactions are closer to the gasoline. They post a price, you select the standardized product, you swipe your debit card, your bank transfers the exact amount to their bank, and everyone is happy. Quite different from the bazaar trade in which you weren't sure of the price or the quality/quantity you were getting, and it was risky just carrying your cash on you.

On the other hand, the modern world has made such transactions the opposite of transparent when looked at from another angle: you don't really have any idea what you are getting or how it got here. How much Nigerian or other blood was spilled bringing that gasoline to market? Were those khakis sewn by kidnapped children in India? How many pesticides and effluent went into the production of your spinach? I'm not going to link examples to each of these or the many other stories we hear on a daily basis. You know them as well as I, perhaps better.

I just thought it was worth pointing out that modern crypto has made the money part of our transactions incredibly secure, so secure that we may soon be able to carry on large swaths of economic activity in cyberspace and outside the surveillance of our insect overlords. However, the actual creation and transport of matter, of things in meatspace, cannot be secured with the intelligent application of prime numbers and collision-free hash algorithms. So, how does one create transparency in the creation of hardware?

There are a few open source hardware movements. The Economist just highlighted some of them (may be a $ link). In it, they mention the Chumby, the Neuros OSD, the RepRap (not mentioned in the article, a comparable project is Fab @ Home), the Tuxphone, OpenMoko, GumStix, and Eric von Hippel's book, Democratizing Innovation. Additionally, I have come across a large number of open source Wi-Fi projects, including this mobile hotspot (but the parts are not OS), this solar-powered grid project to bring the tubes to kids in the developing world (an idea complementary to the $100 laptop), and especially the Linksys WRT54G router. And let's not forget open source automobile projects Oscar, Society for Sustainable Mobility, and c,mm,n (I proposed a framework for how an open source car project might work here).

When you can fire up the matter compiler and build your own car after paying for the matter with a secure transaction, the world is going to be a very different place.

PS: I ran into this article in Wired immediately after posting.
Google is not a search engine. Google is a reputation-management system. And that's one of the most powerful reasons so many CEOs have become more transparent: Online, your rep is quantifiable, findable, and totally unavoidable. In other words, radical transparency is a double-edged sword, but once you know the new rules, you can use it to control your image in ways you never could before.
Food for thought.

Labels: , , , , ,

|

Wednesday, June 04, 2008

Anti-science vs. questionable science

Before my break, over at Crooked Timber, John Quiggin was having a go at John Tierney specifically and Republicans generally on their anti-science stance.

In some of the more hilarious comments, they seem to insinuate that agriculture policies are Republican driven; this may perhaps be the result of a number of non-Americans feeling informed enough to comment on American politics. The most egregious of these policies got their start during the Depression (though some pre-date it), and they grew much larger over the years. That roughly corresponds with the approximately 40-year unbroken period in which Democrats controlled Congress. And as Stephen Downes recently reminded us, the more enlightened Left actually prefers these policies because it maintains order among the Morlocks:

Speaking as someone from the left, I understand the need to provide these subsidies to rural and suburban regions. They are necessary because the free market, left to its own devices, would leave these regions completely unserved.

This would greatly exaggerate the 'time warp' effect, whereby rural regions would be decades behind urban regions, not only in technology, but also education and health care, and ultimately, attitudes and behaviours.

...

It turns out - and we have the empirical evidence for this now - that it is much cheaper to provide subsidies to these regions [rural areas in the US and Canada? Or Africa, Asia, and South America? He seems to have wandered around a bit by this point, so the antecedent is no longer clear - EH] rather than to take a 'law and order' approach. Responding to religious fanaticism, tribalism and the like by war and invasion costs hundreds of billions of dollars - a non-productive subsidy that amounts to thousands of dollars per resident. [I don't think he is still talking about Iowa ... but he does now seem to be implying that invasions are the necessary alternative to foreign aid subsidies! That is a false dilemma.]
Essentially, Downes is admitting that he and his enlightened fellows understand the need to do these things, but Republicans should nevertheless be blamed for actually doing them, and please ignore the history of New Deal farm policy. [1] I wouldn't let either party off the hook on this, though. Republicans took over Congress with the intent of rolling back some of these subsidies, and actually did so for a while (under a Democrat president), but then reintroduced them all under the Freedom to Farm Bill (under a Republican president). The Democratic Congress recently passed another horrific farm policy law with broad-based Republican support, and the silence over at DailyKos is deafening, except to continue to propagate the meme that the beneficiaries of the bill are Republican farmers.[2] As I said at CT, ag policy is non-partisan. Perhaps I should have said bipartisan?

There is a large group of people who tend to be unified by a mindset that is anti-Western, anti-industrial, anti-free-market. Not all share all aspects of this, and not all share the same level of venom, but they exist. At one end of the spectrum, you have the ignorant, violent kids who tore up Seattle and join ELF, who think that Hayduke was a pansy. At the other, you have the reserved lobbyists of the Sierra Club.

People in this group have a model of the world which is reinforced by pessimistic scientific claims. Anything which looks like an indictment of Western, industrial, modern society is immediately accepted on its face because it reinforces their moral views. This confirmation bias, however, goes unrecognized and unacknowledged because of the myriad of other biases that occur when looking back at it introspectively.

One of the biases which makes it difficult to identify past errors is hindsight bias, the tendency to believe that one's predictive abilities are better than they are. We tend to forget bad predictions and to remember good ones. Confirmation bias is the tendency to search for information that matches our preconceptions. Another is the winners' bias, in which we tend to examine hypotheses that actually turned out to be true in order to determine whether hypotheses are more frequently true, a bias which is in part selection bias and in part the Texas Sharpshooter's Fallacy. Selection bias is a distortion created by the manner in which data are collected, in this case resulting from limiting our selection of hypotheses to examine to after the fact rather than before the fact. The Texas Sharpshooter's fallacy is the determination of the hypothesis after the data have been collected. It would be far more constructive if we were to look at all of the hypotheses before the era which was being forecast and to look at which of those were true rather than looking at only the most memorable ones (recall bias?), which are inevitably the ones that eventually come true.

So, for example, we have the historical examples of epicycles, luminiferous aether, and phlogiston. Epicycles were introduced to try to explain the occasional regress of a planet in order to salvage the Ptolemeic or Earth-centric view of the universe. Luminous ether was the medium in which light travelled. Phlogiston was the element which sustained fire. These were important in their day, but are largely unknown today because they were, of course, wrong. These are just a few of the many now-discredited theories of how our world works.

More recently, we had the saccharin scare. Scarcely remembered today, the saccharin controversy was headline news in its day. Having determined that it caused cancer in rats, the FDA wanted to ban it as a carcinogen. They ran into tremendous popular and corporate opposition, since saccharin was the main sugar substitute in diet drinks at the time. They "compromised" by requiring the incorporation of warning labels. Years later, it was determined that saccharin has an effect particular not just to rats, but more specifically to male rats, in a way that does not effect humans. The entire controversy was completely misguided.[3]

Another example is the global cooling scare of the 1970s. It is still well-known that the Earth cooled during the 1940-1970 period, leading to concerns that the trend would continue until we entered a new Ice Age. Pollution was blamed, though it was noted that we are overdue for another Ice Age (in the literature, this is described as the end of the "interglacial period"). Bring this up on a climate change activist website and they will point out that the scare was largely created by articles in Newsweek and National Geographic magazines but was not predicted by scientists. While true that it was brought to the fore of public attention by the popular press, and that few scientists were predicting as opposed to positing the possibility of a new Ice age, the actual history at the time proves my point that these memes achieve some resonance in popular opinion despite the fact that they aren't true.[4] Afterwards, those who believed and advocated strong action claim to have never believed strongly.

It's almost a Lake Wobegon effect: all of today's environmentalists are above average in their ability to have picked only the true environmental scares of the 1970s. How did the belief in global cooling ever get so popular? Was there a die-off among environmentalists in the 1980s? That damn Reagan is probably behind it.

Finally, we have people who accept the claims of Paul Ehrlich:
"The battle to feed humanity is over. In the 1970s the world will undergo famines . . . hundreds of millions of people (including Americans) are going to starve to death." [1968]

"I would take even money that England will not exist in the year 2000." [1969]
Far from being ostracised for making a string (or two) of laughably wrong predictions, Ehrlich has received several awards for his "research". People want so badly to believe him that they continue to discount his way-off-the-mark predictions and accept his newer work at face value.
------------------------------
Most scientists must necessarily be wrong most of the time, at least in published journal articles. This is the nature of science as a search for truth (truth which is as yet unknown): one must submit hypotheses to tests. Either most of those hypotheses must be wrong, or scientists are surprisingly good at guessing right answers, or publication bias is a factor. In fact, those are the findings of scientists researching the results of research: see this article by John Ioannidis and this article by Douglas Allchin for examples.

What is required is a substantial amount of skepticism, even for "accepted" conclusions. At one time in the not-so-distant past, ulcers were thought to be related to nerves, stomach chemistry, and diet. Not until 1979-1981, when two Australian researchers (Warren and Marshall) showed that most peptic ulcers were caused by a bacterium, Helicobacter pylori, did we have the truth. It is fortunate that Warren and Marshall rejected the consensus on this. Howard Aiken's assertion that you shouldn't "worry about people stealing your ideas. If your ideas are any good, you'll have to ram them down people's throats," summarizes the reception Warren and Marshall's ideas received from the medical community. Humans are not good at identifying truth that is at odds with their world view, or at identifying when falsehoods are confirming their world view.

We must be as skeptical of those who claimed that they always knew that Global Warming was true as of Anthropogenic Global Warming (AGW) itself, even if (or perhaps especially if) we believe it to be true. Skeptical, not hateful, dismissive, and/or obtuse. Not skepticism because we "feel" it isn't true, but skepticism for its own sake. The doctors who doubted Warren and Marshall made them prove their claims; that's as good for all of us as is the skepticism of Warren and Marshall that led to the discovery in the first place. Those who claimed they always knew the truth of AGW (especially the non-scientists), even when the evidence was more scant than at present, likely believed (and continue to believe) every pessimistic prediction, but conveniently forgot those that later turned out to be false. They weren't prescient; they were and remain ignorant. They are also ignorant of their ignorance.

---------------------------------------------------------------
[1] A while back, I said that the Left may not be in favor of the Police State, but they are in favor of a Police State. They build it and then feign surprise when the other side uses it. They refuse to believe reports that their own guys use it for anything less than righteousness.

[2] This continues to be a vexing problem about which I intend to post sometime in the near future. That is, a completely misguided application of demographics along with confusing correlation with causation. In this case, we have policies which benefit farmers. Farmers are known to live in rural states. Rural states tend to vote Republican. However, less than 2% of the population farms. Furthermore, it is well known that a small portion of farmers receives most of the subsidies. They alone cannot account for the number of votes received by Republicans. Believing that it is Republicans representing Republican farmers that managed to pass a bill 318-106 in the House, and 81-15 in the Senate, defies explanation on any grounds other than partisan blindness. When you further find out that of those 15 voting against it, only two were Democrats while 13 were Republicans, you really must examine your premises. In other words, it is time for Democrats to drop the sanctimony on farm policy.

In other news (and a demonstration of this same misguided approach to demographics), Democrats are the party of the wealthy and they emit most of the greenhouse emissions. More to come on this, eventually.

[3] Curiously, Quiggin's response regarding the saccharin scare was that it was driven by the USDA, which he sees as a Republican creature:
As regards saccharin, a quick look at Wikipedia reveals that the anti-saccharin push came from USDA. I don’t think it would be too hard to look behind the curtain to red-state sugar and corn producers.
In the first place, the Wiki article specifically points out that the USDA opposition was mostly one man acting in accordance with the law, a law written by the meat-packing industry, so I can't completely discount Quiggin's claim. But Quiggin's assertion that red-state farmers were behind it is typical of the partisan blind under which the subjects of this article labor. Yes, some of those farmers were Republican. Some are Democrat, but he is blind to them. The laws were all written, supported, and not repealed by the Democrats even though they have controlled Congress for most of the period since 1907, and the White House for about half of it.

In the second, this doesn't explain how the FDA came to attempt to ban it.

In the third, doesn't this illustrate exactly what many of us have been saying with respect to regulatory capture? Specifically, that the government mostly exists to defend corporations in the guise of defending the average citizen?

[4] In fact, this has become a new interest of mine: How do such ideas get created and transmitted to seats of power? It isn't always via the press, and the locus of power is not always popular opinion. Take, for example, the ideas of the German Historical school, which got mainstreamed under the name of Progressivism by a route that seems to have included Robert Ely, John Commons, Herbert Croly, Robert M. LaFollette Sr., Teddy Roosevelt, Louis Brandeis, and finally FDR's cabinet. Why that school of thought? Why that route? Pure chance?

My point here repeatedly escapes the comprehension of the pessimists. I am not saying, "Scientific consensus in the 1970s was in favor of global cooling and impending an Ice Age." I am saying, "Public opinion in the 1970s was tipping toward a belief in an impending Ice Age and a desire to do something about it." It therefore doesn't matter what scientists were publishing in journals, or that there was no scientific consensus predicting the end of the interglacial period. What matters is that a significant number of voters believed it. How did they receive the information? How credible did they perceive it to be? How strong were their beliefs, and how far were they willing to go to act on them?

Labels: , , ,

|

Tuesday, June 03, 2008

Transparency

So one of the outcomes of the unfortunate series of events from my last post has been a discovery of Stephen Schneier's excellent blog, Schneier on Security, and a subsequent gaffe on my part. In one of his posts, he linked to an article written by Naomi Klein for Rolling Stone in which she warns of the panopticon currently being built in China with technology being developed by defense contractors in the US. My gaffe was to react to the author's identity and to scan the article for the earmarks of a Naomi Klein piece: half-truths and wild assertions. I found my quarry: first, in that she refers to China, a police state if there ever was one, as an example of a free market, and second, that the alarmist article is riddled with examples of ongoing criminal activities in this totalitarian world, a paradox or irony which she fails to note. Besides those two drawbacks, I find that her alarmism may not be without foundation.

Which is one of the reasons why I find comments like Bernard Yomtov's comment in this thread so meaningless. Yes, there are probably libertarians (Kevin Carson's "vulgar" libertarians) who are more interested in lower taxes and repeal of motorcycle helmet laws than they are in issues like indefinite incarceration on the basis of secret evidence. But the libertarian community with which I am most familiar (a) does not suffer from that affliction, and (b) is more likely to see civil rights as a continuum in which the differences are of degree as well as of kind [1]. Tune in to Radley Balko's blog for a few weeks, see kids arrested and choked for skateboarding, see no-knock drug-war raids turn into murder of innocents, see kids taken from their parents on the slightest pretexts (see below).

In related articles, I see that Kevin has a similar reaction to Naomi in this essay reviewing her The Shock Doctrine [2], and I also see where Schneier makes a good point regarding transparency in this article in which he reviewed David Brin's The Transparent Society, revived on its 10th anniversary. He says,

If I disclose information to you, your power with respect to me increases. One way to address this power imbalance is for you to similarly disclose information to me. We both have less privacy, but the balance of power is maintained. But this mechanism fails utterly if you and I have different power levels to begin with.

An example will make this clearer. You're stopped by a police officer, who demands to see identification. Divulging your identity will give the officer enormous power over you: He or she can search police databases using the information on your ID; he or she can create a police record attached to your name; he or she can put you on this or that secret terrorist watch list. Asking to see the officer's ID in return gives you no comparable power over him or her. The power imbalance is too great, and mutual disclosure does not make it OK.

Which brings me to one of the points of interest I have had with respect not only to corporate security policy, but to large organizations in general: one of the basic problems is the establishment and maintenance of transparency. How are decisions arrived at? Who has the decision authority? What is the basis of a decision? How may the decision be appealed?

This is the genius of Open Source: you don't like it, ultimately, you could change it yourself. But more importantly, Open Source embraces Eric Raymond's reformulation of Claude Shannon's reformulation of Kerckhoffs' principle: a system that is based on the assumption that the enemy -- who could be MegaSoft or the state -- has the code (but not the key) because the code is freely accessible (and therefore subject to close scrutiny) is more secure than a system whose operation is a closely held secret. This same principle would seem to apply to anything that someone would want to do, whether a company or a state, whether developing code or a new chemical or passing a farm bill or budget.

Another aspect of transparency: authority and responsibility must lie at the same locus. Remove one and leave the other and you end up with an irresponsible despot and/or powerless bureaucrat. Most large organizations are filled with both, sometimes the same person acting at different times in different capacities. That is what lies at the bottom of one of the threads undertaken in this post: a mistake by a parent leads to the detention of the child even though everyone at every stage acknowledges it is the wrong thing to do (original story here). In that case, the system is transparent to neither the victims nor to the perpetrators, the latter who don't even know how to appeal a decision they would like to appeal but nevertheless feel compelled to enforce.

Transparency seems more easily achievable in smaller organizations and in less hierarchical organizations. It isn't a slam dunk: presenting a grievance to a single individual with authority may get you further at less cost than attempting to convince the majority of a democratic body. This was the subject of Madison's Federalist X essay: the tension between the need to thwart the "violence of faction" while maintaining some semblance of democracy. It is also the basis of the Public Choice theory in The Calculus of Consent. Given the arguments I made in Local, Action: Issues of Scale, I am claiming a close correlation between transparency and communication and the difficulties of achieving either in large social structures.

-----------------------
[1] As to the difference in kind: some civil rights are undermined in reaction to criminals who harm others, while others are undermined in reaction to citizens who might harm themselves. But it seems to me that each side of the spectrum likes to take their favorite pet peeve of the latter type and turn it into one of the former. Thus, for Bill O'Reilly, drugs should be illegal for adults because they eventually end up abusing their children because of the drugs. For the anti-smoking crowd, Second-Hand Tobacco Smoke was a godsend. And now, global warming is the trump card.

It makes sense that if we do not want the police state that we have, then the first laws we should get rid of are those that purport to defend us from ourselves. If we can't even take these off the books, how can anyone believe that we will be able to take laws off the books (or practices off the table) that supposedly protect us from supposedly bad men? Both repeals would require an admission that the state cannot protect us from all risks. After all, they will point out, 19 men with box cutters very nearly managed to kill tens of thousands of people rather than "just" 3000, and going back to that state of affairs invites another such disaster.

Note that that argument is not my argument, but rather the argument you're going to get from the "protect the kids" crowd. It isn't easy to dismiss since it appeals to our sense of personal security; its most recent manifestation was not a Bush speech, but Hillary's "3 AM call" campaign ads. I would argue that we are no better off under the new state of affairs, under which the FBI and TSA are effectively committing a low level type of terrorism. Yes, they may be catching some bad guys and preventing some bad things in the near-term, but they are creating more terrorists abroad. As they keep pointing out to us, preventing acts of terror is a matter of dismal statistics; the bad guys only have to get lucky once, while the "good" guys have to get lucky every time. Creating more future terrorists who only have to get lucky once seems like more rather than less risk, but the alternative is to have an adult conversation with the public about terror.

The Left punted on that. They chose instead to accuse the Bush Admin of "failing to connect the dots" and being "asleep at the switch". Those accusations contributed to the overwrought response and subsequent invasion of Iraq. They should have asserted that this was an unfortunate event requiring in-depth analysis rather than bold action. But the problem is that winning debate points is far more important to the Left than preventing the growth of the police state, which they rather enjoy when they control it.

[2] I cannot follow Kevin down the pro-Chavista path. Not only has Chavez shown autocratic tendencies (extending his own power, shutting down dissenting media), but if the FARC cache is authentic, he has shown a willingness to support violence to undermine neighboring democracies. Yes, Colombia's democracy is tenuous and questionable, but no less than Chavez' own. When Kevin says,
Quite frankly, if my only choices are corporate liberalism and social democracy, and a banana republic on the neoliberal model, I'll take the former any day. If I get to choose between the paternalism of Brave New World and the jackboot in my face of 1984, it won't take me long to decide. I'm not ashamed to say that if my only choices are the welfare statist and neoliberal versions of statism, I'll take the kind of statism whose yoke weighs less heavily on my own back.
he is, by his own tacit admission ("if my choices are ..."), offering a false choice. Let us hope we have at least a third choice.

Labels: , , ,

|

Monday, June 02, 2008

Update - at last!

Geez, so I haven't updated this thing since April? Okay, so I've been busy.

The last posts I started to work on but never completed were a review of a variety of dystopian novels and films I had recently experienced and an exploration of misguided demographic findings of correlation. But from there, I got side-tracked onto a couple of projects that I have put off for a while, and those resonated with things that have been going on in my professional life. For one, I finally got around to renewing an interest in linux. My laptop is now a dual-boot system.

The first time I tried it out many years ago, it was still at the "linux is so cool, if you can't find a driver you can write your own and you can even rewrite and recompile the kernel if you want to!" stage. Sorry, I want to use the computer for other ends, not as an end in itself. Open Source is almost there, perhaps the latest release of Ubuntu (Hardy Heron) puts it over the top. I see where even Dell is selling a laptop with Ubuntu installed.

The next thing was to finally read some Neal Stephenson right after Richard Feynman's Surely you're joking, Mr. Feynman. After plowing through Cryptonomicon, I went down to the local book exchange and pulled off the entire shelf, consisting of Snowcrash, Diamond Age, Zodiac, and Quicksilver. Unfortunately for me, one result of these (especially Feynman and Cryptonomicon) was to resurrect a childhood interest in cryptography.

The interest in crypto has tied in to an unfortunate but ongoing relationship with computer security policy. Unfortunately, the more you learn about this type of thing, the more paranoid you become. Yeah, having to memorize lots of strong passwords is difficult, so difficult that many people resort to writing them down and storing that within arm's reach of their computer, but without those and other seemingly overkill measures, you might as well be running a wide-open system. But then again, even if you have the strongest crypto around, the weakest link is usually the human factor, as illustrated by the common knowledge that most password's are within arm's reach of their computer. Comcast recently learned this lesson the hard way. And while it may be hard to pity Comcast, anyone who has read Cliff Stoll's The Cuckoo's Egg should understand that hacking pranks like these create a sense of alarm and lost innocence, sorta like when the residents of a small town have to start locking their doors because someone thought it would be fun to vandalize one house with unlocked doors. But this hardly excuses the overbearing stance of corporate security, who would really rather that you didn't have a computer at all.

Speaking of small towns, photos from our recent visit to Lake Wobegon.

Labels: , , , ,

|