Tuesday, January 16, 2007

Visualizing Information

(credit Bayesian Heresy). I’ve taught lectures on information and data visualization. Next time I’ll include this one:

New Economist: 98 ways to visualise data

... Let's start with the visualisation of data. Aleks Jakulins at the Statistical Modeling, Causal Inference, and Social Science blog points us to a stunning periodic table of visualisation methods. The table provides 98 methods of visualising information. ...

Humanity: up from the sludge

Arnold Klink begins a review of humanity’s improvements with a Steven Pinker quote:

TCS Daily - Appreciating Our Moral and Mental Development

... "In 16th century Paris, a popular form of entertainment was cat-burning, in which a cat was hoisted on a stage and was slowly lowered into a fire. According to the historian Norman Davies, "the spectators, including kings and queens, shrieked with laughter as the animals, howling with pain, were singed, roasted, and finally carbonized." ...

Oh.

Elsewhere I recently read the claim that the male violent death rate in a hunter-gatherer society is 30%. Our violent death rates are rather lower.

Two thousand years ago a bizarre doctrine that called for the forgiveness of enemies actually almost got off the ground. Fragments of it remain today. So, it’s not linear progress, but it’s hard to deny that it’s progress.

But why? That’s the interesting question. Is it all cultural? Is there some biological component as well? I suspect there’s something biological. I wonder if we have “aggression set points” that we switch between based on early childhood experience and maternal prolactin secretion. I wonder if, on a longer timeframe, we have “backup-gene collections” (these apparently exist in many species) that allow a human population to effectively switch cooperation patterns in a few generations depending on changing environments …

Update 1/17/2007: We know that mammals are capable of insect-like social systems (naked mole rats) and of differential adult behaviors based on early experiences. We know humans are among the most socially integrated of all mammals, rivaling the social integration of the insects. We also have "Hellstrom's Hive", a 1970s science fiction short story series that compared human social organization to insect life. I believe many insect colonies alter colony behavior and the drone/warrior balance based on environmental conditions. Hmmm.

Google blogger and the problem of asymmetric relationships

Google’s blogger hosts about 10 of my blogs, including this one. It has been, by an order of magnitude, the most troublesome of the Google tools and services that I’ve used. Recently Google has been moving blogs from the old infrastructure, which had been fairly trouble free over the past six months, to a new “better” (now buggier) infrastructure. The transition has not been going well; my blogs are too large and complex to move and they currently live on the increasingly abandoned old servers.

All of which has led to some thought about why this relationship isn’t working out. I think it’s a slightly different class of problem form the one that I use to have with Gmail. From my post to a Google Group:

Google Groups : Blogger Help Group > Something Is Broken

... Blogger (and Google) illustrates the generic problem of an asymmetric relationship between customer and vendor. Blogger may produce income for Google, or help tune searches, but the vast majority of Blogger blogs individually produce a trivial amount of revenue….

That would be fine if we and Google held similar opinions of the value of our individual blogs. If we both felt they were of trivial importance our relationship would work. Alas, we may value the content we've produced far, far more than Google values it. The relationship is thus doubly asymmetric; we bring little individual value to Google but we value our work far more than Google does.

The lesson I've drawn from my Blogger experiences over the years is not that Google is particularly evil, it's rather than asymmetric relationships are highly problematic. (Obviously the issue is systemic and applies to personal as well as business relationships.) If and when I move from Google/Blogger, it won't be to another asymmetric relationship. I would only move to a vendor where my voice mattered. Since my blogs will never generate much revenue, this means I will be paying for services in cash. ...

One part of the solution is a level of indirection between URL and blog service — at the very least one must control the URL. The value equation may change, and that may require a change in blog service …

Smoking: now for the menthol

My wife and I held our most recent weekly family meeting in the pub (no, the kids were home). It had been a long week.

On the way home, we passed the smokers in an external wind shelter. It was 10 F or so; they wouldn't have looked good in an ad. That's when we remembered pubs, bars, even restaurants (classrooms once upon a time) used to have smokers in them. We've gotten quickly used to a smoke-free world in the Twin Cities; a state wide ban is on the way.

It's not just the “progressive states”. When I was studying french in rural Quebec, my host family would meet to roll cigarettes every night (a most pleasant meeting too). Quebec now has a provincial smoking ban.

What's next? I usually think about regulating nicotine content, but an Atlantic (paywall) "primary sources" post reminded me that menthol is an adjuvant agent of addiction -- it slows nicotine metabolism. Eliminating menthol may be the logical next step ...

Update 1/16/07: Simple regulation would be wasteful of course. The better approach would to first put a differential tax on mentholated cigarettes. After a few years, add differential taxes based on nicotine content. This is the equivalent of a progressive decrease in nicotine patch strength done on a nationwide basis.

Monday, January 15, 2007

Dale Carnegie - the condensed version

After the Great War and before WW II, following a divorce and during the Great Depression, Dale Carnegie wrote an optimistic book - "How to win friends and influence people".

It's good bathroom reading, easiest to take a little bit at a time. On reading it I wonder if some of Carnegie's "friends" came to doubt the sincerity of his sentiments, but it's hard to argue with the general principles of the latter half of the book.

I figured it'd be useful to have an edited version of his "rules" at hand, and thanks to Wikipedia that's not hard. These are the "rules" I like the most ...
How to Win Friends and Influence People

Twelve Ways to Win People to Your Way of Thinking

If you're wrong, admit it quickly and emphatically.
Begin in a friendly way.
Start with questions the other person will answer yes to.
Let the other person do the talking.
Let the other person feel the idea is his/hers.
Try honestly to see things from the other persons point of view.

Nine Ways to Change People Without Giving Offense or Arousing Resentment

Begin with praise and honest appreciation.
Call attention to other people's mistakes indirectly.
Talk about your own mistakes first.
Ask questions instead of giving direct orders.
Let the other person save face.
Praise every improvement.
Give them a fine reputation to live up to.
Encourage them by making their faults seem easy to correct.
Now in the modern world, try doing all this over the phone, without ever seeing the other person!

PS. I want anyone who might know me to understand I'm not delusional -- I don't claim to practice these recommendations. Still, one must have aspirations...

How to crack a system

Cryptonomicon, a reasonably popular science fiction book, includes an excellent introduction to cryptography. So I actually knew most of what Schneier tells us about how passwords are cracked...
Schneier on Security: Choosing Secure Passwords

... AccessData sells another program, Forensic Toolkit, that, among other things, scans a hard drive for every printable character string. It looks in documents, in the Registry, in e-mail, in swap files, in deleted space on the hard drive ... everywhere. And it creates a dictionary from that, and feeds it into PRTK.

And PRTK breaks more than 50 percent of passwords from this dictionary alone.

What's happening is that the Windows operating system's memory management leaves data all over the place in the normal course of operations. You'll type your password into a program, and it gets stored in memory somewhere. Windows swaps the page out to disk, and it becomes the tail end of some file. It gets moved to some far out portion of your hard drive, and there it'll sit forever. Linux and Mac OS aren't any better in this regard...
It's somewhat reassuring to know Schneier isn't perfect. As others note in the comments (I looked) he's wrong about OS X -- both the user directory and the swap file can be optionally encrypted with modest performance impacts.

Personally, I store my passwords in an encrypted database and I generate the important ones from GRC's password generator. My answers to the "personal question" are long strings of angrily typed characters ...

XP, however, is kind of hopeless. Maybe Vista is different.

Seeking Dickens: mental illness and prisons

We really need a Charles Dickens for the 21st century. Emphases mine.
The Mentally Ill, Behind Bars - New York Times
Bernard E. Harcourt, a professor of law and criminology at the University of Chicago, is the author of “Against Prediction: Profiling, Policing and Punishing in an Actuarial Age.”

... Over the past 40 years, the United States dismantled a colossal mental health complex and rebuilt — bed by bed — an enormous prison. During the 20th century we exhibited a schizophrenic relationship to deviance.

After more than 50 years of stability, federal and state prison populations skyrocketed from under 200,000 persons in 1970 to more than 1.3 million in 2002. That year, our imprisonment rate rose above 600 inmates per 100,000 adults. With the inclusion of an additional 700,000 inmates in jail, we now incarcerate more than two million people — resulting in the highest incarceration number and rate in the world, five times that of Britain and 12 times that of Japan.

What few people realize, though, is that in the 1940s and ’50s we institutionalized people at even higher rates — only it was in mental hospitals and asylums. Simply put, when the data on state and county mental hospitalization rates are combined with the data on prison rates for 1928 through 2000, the imprisonment revolution of the late 20th century barely reaches the level we experienced at mid-century. Our current culture of control is by no means new.

... It should be clear why there is such a large proportion of mentally ill persons in our prisons: individuals who used to be tracked for mental health treatment are now getting a one-way ticket to jail.

Of course, there are important demographic differences between the two populations. In 1937, women represented 48 percent of residents in state mental hospitals. In contrast, new prison admissions have consistently been 95 percent male. Also, the mental health patients from the 1930s to the 1960s were older and whiter than prison inmates of the 1990s.

... One of the most reliable studies estimates that the increased prison population over the 1990s accounted for about a third of the overall drop in crime that decade.

However, prisons are not the only institutions that seem to have this effect. In a recent study, I demonstrated that the rate of institutionalization — including mental hospitals — was a far better predictor of serious violent crime from 1926 to 2000 than just prison populations. The data reveal a robust negative relationship between overall institutionalization (prisons and asylums) and homicide. Preliminary findings based on state-level panel data confirm these results...

Harcourt is careful to note that the prison/institution relationship is not a simple substitution, but I'd be surprised if there weren't a strong relationship. Humans are really not all that well put together. We're running a 15,000 yo cognitive system way out of its operational range. We do astoundingly well all things considered -- but, really, we're very buggy thinkers. We need to to rethink the "problem of the weak" on many levels.

Damn Interesting: The Woman with a Limp

Damn Interesting is a new pickup for me. I can't recall how I came across it. It's well named; it belongs on everyone's bloglist. Enjoy!

Domestic spying: thank heavens for the Dems

An old story, with a new twist. The Pentagon (you know, the military?) has been using "national security letters" to obtain banking records on "suspicious" US citizens. Cheney declares no patriot would object.

Once this would have been a bit shocking, but it's a yawn. The US military running its own domestic spy operations? Hardly a surprise! This time, however, there's a twist ...
Cheney Defends Efforts to Obtain Financial Records - New York Times

...Representative Silvestre Reyes, a Texas Democrat who is the new chairman of the House Intelligence Committee, said his panel would examine the matter. Mr. Reyes also indicated that he might renew efforts to pass a law requiring various agencies to get court approval before issuing national security letters.
We're much better off than we were a month ago ...

Dubai: Slate's well-traveled tales

Looking for mammon in the Muslim World (By Seth Stevenson) is the first in a series of travel essays on Dubai, a small nation with vast wealthy and curious ambitions. It's fun travel writing, and a great story -- five entries so far. Recommended, of course.

Pasteurized milk, gut ecology, and google scholar

I left the milk out a bit long yesterday. It's still good, but the wee ones are at work, we'd better drink it quickly.

Which led me to think about how the pasteurization or irradiation of milk has altered the ecology of the human gut. Now that we think of ourselves as a superorganism, a walking ecology, it's obvious that altering gut ecology is not necessarily a good thing. It could impact obesity rates, bowel disorders, etc.

Now, of course, whacky ideas like this can be quickly researched. I didn't see an obvious directly relevant article, but it looks like the domain is being reasonably well explored. I also see that Google has quietly done some nice work with Google Scholar...

Sunday, January 14, 2007

Books about Minnesota: Why they annoy me

Virtually all books about Minnesota annoy me, and it's getting worse. Take this seemingly excellent book: Hiking Minnesota With Kids. What could be wrong with that?

For one thing, most Minnesotans live within a one hour drive of the Wisconsin border. I don't want a "Hiking Minnesota with Kids" book, I want a "Hiking Minnesota and Wisconsin with Kids" book, or a "Hiking within 300 miles of the Twin Cities book".

I'm going to start boycotting all MN or WI only books, and buy every MN/WI book I see.

I can't be the only TC resident in a snit about this ...

PS. Some of the best hiking and around the TCs is in western Wisconsin ...

Even the dimmest eyes are opening ... A right wing flack reneges

Shrillblog quotes Rob Dreher (no permalink, alas). I've never heard of him, he seems like a dim right wing pundit. Things are very bad for Bush, and, alas, for all of us, when someone like him wakes up...

Crunchy Con: My All Things Considered commentary - Rod Dreher, Conservative blog, Beliefnet conservative politics and religion blog:


... I talk about coming to terms with the end of an illusion. As someone who came of political age under Reagan, I've been a conservative for most of my life (for the sake of brevity, NPR edited out the part of the essay in which I explained that I'd had a high school and early-college dalliance with liberalism). I disdained the Vietnam-era "hippie" mentality with regard to national security. I took it for granted that those people were hung up on Vietnam, and ought not be listened to because they were blame-America-first liberals....

I formed my political views on national security in the confident glow of Reaganism. For me, it was a fact of life that Republicans were strong, capable and confident, and Democrats were weak, vacillating and incompetent.... When Bush led us into the Iraq War, I thought the liberals who predicted doom -- and, crucially, the conservatives (like Buchanan) who did as well -- were either fools, cowards or unpatriotic. But now I see that I was the fool. In the NPR piece, I wrote about how I sat there watching Bush's speech and thought that when they get old enough to understand these things, I have got to teach my children never, ever to take the word of presidents or generals at face value. To question authority, because the government will send you off to kill and die for noble-sounding rot (e.g., crusading for democracy in the Middle East). And it hit me that this is precisely the message that so many of those who lived through the Vietnam experience tried to tell my generation -- in my case, and in the case of so many other Gen X Reagan Youth, in vain.

I give him credit for speaking up now. It may yet do some good it if turns enough GOP senators to prevent a filibuster of moves to reduce Bush's power.

Thursday, January 11, 2007

Blumenthal on the end-game of the "surge"

How it unfolded, and how Baker failed to turn the tide.

Salon.com | Shuttle without diplomacy

… Informed correspondents of the Washington Post and New York Times related in conversation that Bush furiously called the [Iraq Study Group] report "a flaming turd," but his colorful remark was not published. Perhaps it was apocryphal. Nonetheless, it conveyed the intensity of his hostile rejection....

... Donald Rumsfeld had been sacrificed as the secretary of defense, but his replacement, Robert Gates, a former director of the CIA and member of the ISG, turned from skeptic into team player. The Joint Chiefs of Staff; Gen. John Abizaid, head of Central Command; and Gen. George Casey, commander in Iraq, all opposed the "surge" as no answer. Cheney and the neocons saw their opposition as the opening for purging and blaming them. The Joint Chiefs were ignored and sidelined, Abizaid was forced into retirement and Casey was removed (sent into internal exile as Army chief of staff). Their dissent, leaked to the Washington Post for appearance in the paper on the day of Bush's "surge" speech, was an extraordinary gesture by the senior military leaders to distance themselves from impending failure.,,

… The State Department has been completely sidelined in the making of Bush's latest and last policy on Iraq. Its experience in the Balkans remains thoroughly ignored. And Rice does nothing to call it to Bush's attention, for that would require her to point out his shortcomings. The State Department founders like a ghost ship. Rice meanders back and forth to and from the Middle East, the shuttle without the diplomacy.

After twice rejecting the job of deputy secretary of state, John Negroponte, the director of national intelligence, was implored to accept it. In exchanging a Cabinet post for a sub-Cabinet one, a position of policymaking for an administrative post, Negroponte excited rumors that he would only have decided to make the switch if he believed that Rice would eventually leave and he would ascend to her job. But, once again, the logic of that Washington gossip is merely rational. Rice the irrelevancy remains Bush's indispensable devotee.

Congress alone will need to stop this president.

Sarbanes-Oxley means no features in future software updates from publicly traded companies?

The claim is that Sarbanes-Oxley has the unintended consequence of making it illegal, or costly, to distribute ‘free’ software updates.

Oh, about that 802.11n card in your C2D Mac | iLounge

... I’m not going to claim to understand this next part, which really just makes no sense to me at all, but the claim Apple’s making is that it _can’t_ give you the 802.11n-unlocking software for free. The reason: the Core 2 Duo Macs weren’t advertised as 802.11n-ready, and a little law called the Sarbanes-Oxley Act supposedly prohibits Apple from giving away an unadvertised new feature for one of its products. Hence, said the Apple rep, the company’s not distributing new _features_ in Software Update any more, just _bug fixes._ Because of Sarbanes-Oxley. If this is an accurate statement of Apple’s position, which as an attorney (but not one with any Sarbanes background) I find at least plausible, this is really crazy. ...

Presumably this falls under the category of a publicly traded company cheating shareholders of revenue. It is certainly startling if true, but it does seem plausible. It may, of course, be something that vendors like. It would not apply to non-publicly traded companies, which produce most of the software I like.

Sarbanes-Oxley is due for a review by Congress. I’ll ask my Representative to take a look at this claim.

Update 1/18/07: It looks like this is genuine:
Apple's alleged 802.11n enabler fee: blame Enron etc. | Reg Hardware

... The reason: changing a product's functionality changes the timeframe in which the manufacturer can recognise revenue from the sale of that product.

If Apple begins selling a product at the start of, say, Q1 and then adds a previously unadvertised feature to it at the beginning of Q2, under Sarbanes-Oxley, it has to recognise revenues from the product from Q2 onwards. Revenues recognised in Q1 run contrary to the rules enshrined in the Act. Unless, of course, the extra feature is a 'new' product, attracting its own revenue stream.

Crazy, but that's accountancy for you. Had Apple actually said at their launch that its latest Macs would one day be upgradeable to 802.11n, it could have avoided the charge, it seems...

Microsoft Vista, it was announced today, will be installed with a comprehensive feature set which may be unlocked over time for a fee. I wonder if Sarbanes-Oxley determined that.

I'm not sure this is all bad. Companies now have a clear economic incentive to deliver incremental value to existing solutions. If and when I want 802.11n support I have no problem with paying $5.00 for a driver. Perhaps Apple will consider adding phone support to iSync for a similar fee; I'd be delighted to pay $5.00 for official iSync support of my Motorola V3M.

Update 3/10/07: I'd read some coverage that claimed Apple was interpreting Sarbanes-Oxley incorrectly. I'd written our representative to ask about this, and Betty McCollum's office replied "Apple has to account for the separate value of a software upgrade that allows for additional capabilities from the hardware.... a nominal fee ... establishes a reportable value for the upgrade." So Apple has interpreted the law as congress understands it. At least when it comes to enabling new hardware capabilities, SO means Apple must account for the value delivered. A nominal fee is one way to do that. I wonder if there's more wiggle room when no hardware capabilities change ...?