Saturday, March 28, 2009

Carrington class solar storms -- another reason to own a cabin with running water

(Via KW) We're used to estimating the frequency of calamities by studying the historic record.

That works for meteors, earthquakes, climate change, volcanoes and the like.

It doesn't work for Carrington class solar events, because they didn't mean much until recently ...

Space storm alert: 90 seconds from catastrophe - space - 23 March 2009 - New Scientist

IT IS midnight on 22 September 2012 and the skies above Manhattan are filled with a flickering curtain of colourful light. Few New Yorkers have seen the aurora this far south but their fascination is short-lived. Within a few seconds, electric bulbs dim and flicker, then become unusually bright for a fleeting moment. Then all the lights in the state go out. Within 90 seconds, the entire eastern half of the US is without power.

A year later and millions of Americans are dead and the nation's infrastructure lies in tatters...

... an extraordinary report funded by NASA and issued by the US National Academy of Sciences (NAS) in January this year claims it could do just that....

...The surface of the sun is a roiling mass of plasma - charged high-energy particles - some of which escape the surface and travel through space as the solar wind. From time to time, that wind carries a billion-tonne glob of plasma, a fireball known as a coronal mass ejection (see "When hell comes to Earth"). If one should hit the Earth's magnetic shield, the result could be truly devastating.

The incursion of the plasma into our atmosphere causes rapid changes in the configuration of Earth's magnetic field which, in turn, induce currents in the long wires of the power grids. The grids were not built to handle this sort of direct current electricity. The greatest danger is at the step-up and step-down transformers used to convert power from its transport voltage to domestically useful voltage. The increased DC current creates strong magnetic fields that saturate a transformer's magnetic core. The result is runaway current in the transformer's copper wiring, which rapidly heats up and melts. This is exactly what happened in the Canadian province of Quebec in March 1989, and six million people spent 9 hours without electricity. But things could get much, much worse than that...

... The most serious space weather event in history happened in 1859. It is known as the Carrington event, after the British amateur astronomer Richard Carrington, who was the first to note its cause: "two patches of intensely bright and white light" emanating from a large group of sunspots. The Carrington event comprised eight days of severe space weather.

There were eyewitness accounts of stunning auroras, even at equatorial latitudes. The world's telegraph networks experienced severe disruptions, and Victorian magnetometers were driven off the scale.

... According to the NAS report, a severe space weather event in the US could induce ground currents that would knock out 300 key transformers within about 90 seconds, cutting off the power for more than 130 million people (see map)...

... The truly shocking finding is that this whole situation would not improve for months, maybe years: melted transformer hubs cannot be repaired, only replaced. "From the surveys I've done, you might have a few spare transformers around, but installing a new one takes a well-trained crew a week or more," ...

... The good news is that, given enough warning, the utility companies can take precautions, such as adjusting voltages and loads, and restricting transfers of energy so that sudden spikes in current don't cause cascade failures. There is still more bad news, however. Our early warning system is becoming more unreliable by the day.

By far the most important indicator of incoming space weather is NASA's Advanced Composition Explorer (ACE) ...ACE can provide between 15 and 45 minutes' warning of any incoming geomagnetic storms. The power companies need about 15 minutes to prepare their systems for a critical event, so that would seem passable...

...observations of the sun and magnetometer readings during the Carrington event shows that the coronal mass ejection was travelling so fast it took less than 15 minutes to get from where ACE is positioned to Earth...

... ACE is 11 years old, and operating well beyond its planned lifespan...

What we need is a technological fix that helps with some common electrical system problem -- so we can justify it that way -- with a side-effect of protecting against Carrington-class solar storms. Preferably something that doesn't rely on an early warning system or human judgment, but behaves more like the household Ground Fault Interrupt.

Reinhardt on Obama's health plan

Uwe Reinhardt is a Clinton economist who was one of the godfather's of Hilary's failed reform plan.

Today he gives us a high level outline of Obama's reform plan and promises more to come.

Sure looks hopeless to me. I can't line it up in any significant way with what we have to do.

I'd be happy if all they did was rip health related benefits out of employment -- which was actually a McCain position.

How big should financial firms be?

The Baseline Scenario suggests we simplify financial services regulation by setting size limits. Then, in theory, market forces can punish the foolish -- without destroying the economy.

I'm sympathetic.

The bit of the post I liked best, however, was the reference to a classic paper on corporations ...
Big and Small The Baseline Scenario

.... this is the issue that Ronald Coase discussed in “The Nature of the Firm” (Wikipedia summary; paper). A firm’s optimal size is reached when the transaction costs of doing business in the market equal the administrative costs of managing the firm; the bigger the firm, the higher the administrative costs... To this equation, we now need to add the social costs (negative externalities) of being Too Big To Fail: moral hazard, socialized losses, and so on.
I suspect a simple interpretation of this theorem would find many corporations are far larger than the "optimum".

What the theorem may miss (I've not read the paper) is that corporations don't just compete in the marketplace. They attack each other through acquisitions, cutting off suppliers, buying up critical components, purchasing senators (typically by paying for their election and reelection) and so on. In other words, within certain boundaries, corporations go to war.

That may favor bigger corporations than simplistic models of economic efficiency.

Friday, March 27, 2009

Death to Captchas

Beck when machines had trouble solving simple Captchas, they weren't a bad idea.

Now the machines are much better at solving them than we are. I see red when I see a Captcha, no matter the color of the cursed thing...

Recent Stuff That Bothers Me - Pogue’s Posts - NYTimes.com

... These days, blogs and Web sites often require you to prove that you’re human by typing in the text version of some distorted picture of a word. The idea is to screen out automated software spambots that fill the Comments area with auto-generated ads...

Captcha

... I suddenly realized how much I hate these things when I got a note from reader Jason Donovan, who’s started a Web site where you can post your favorite (meaning most ridiculous) Captcha images.

Some of the starter images posted there aren’t hard to figure out. But the ones in color, one of which I’ve pasted here, are living, breathing proof that these things have gotten quite out of control.

I moderate all comments and foreswear the cursed Captcha. It was a nice try, but the experiment failed. The machines aced the Turing test.

Thursday, March 26, 2009

Journalists braver than you or I

I hope my sister-in-law is not reading this …

Out of Africa: The final dispatch | World news | The Guardian by Chris McGreal

They were the best of times in Africa, and the worst. They were the years when South Africa was swept away by the belief that it was a nation blessed, a moral beacon to the world, symbolised by a single moment as Nelson Mandela stood outside a small KwaZulu school in April 1994, dropped his vote into the ballot box with a cross next to his own name, and undid what an entire system had been constructed to prevent…

… Weeks after watching Mandela vote, I was standing at a church among thousands of corpses rising from the ground…

…I had met the man responsible for all this a few hours earlier. ClĂ©ment Kayishema was a doctor and, at one time, head of the local hospital, but by the time of the genocide he was a political force as the governor of Kibuye province. When I turned up in his town, he directed that I be held in a hotel-turned-barracks. One day he would have cause to regret that…

…after two decades of watching failed leadership, the Africans that have made the greatest impression on me are the extraordinary individuals who stood against that tide.

In South Africa there is Zackie Achmat, an HIV-positive gay Muslim man of Indian extraction and ANC member, who led the campaign against Mbeki's perverted denial of life-saving anti-Aids drugs to poor black people…

…the women in eastern Congo who venture into the most dangerous areas to rescue other women from years of systematic mass rape by the gangs of armed militias that amount to the only form of authority over vast territories. Or the Nigerian journalists who risked assassination or long sentences in hellish prisons to expose the truth about the military dictators plundering their country…

…And there are those who names cannot, for now, be revealed. They include the Zimbabwean doctors who have for years lived with the risk of arrest, torture and even death to run an underground railroad to help the victims of Mugabe's sustained and bloody terror against his people…

…Sosthene Niyitegeka. The Hutu shopkeeper and pastor risked everything - his own life and that of his wife and children - to save every Tutsi in his village at the height of the Rwandan genocide

...A decade later, Mbeki's failed leadership is principally remembered for sacrificing the lives of hundreds of thousands of people while he fiddled around in league with a group of maverick scientists who questioned the causes of Aids and the established methods of keeping HIV-positive people alive. Mbeki's blocking of the life-saving drugs to millions of people was his greatest crime, but his stature was further eroded abroad by his malign manipulation of Zimbabwe's political crisis to help keep Mugabe in power. When he wasn't squandering South Africa's moral authority over Zimbabwe, Mandela's successor was wasting it at the UN security council protecting Burma's military regime...

McGreal’s stories, including his own, are awe inspiring, but not examples you necessarily want your loved ones to follow …

The crash began in 1984?

After 1984 financial sector compensation rose much more quickly than average compensation (The Baseline Scenario). Even if one doesn’t buy all of the associated reasoning the chart deserves an explanation.

The companion Atlantic Article is online early with much more detail. The basic premise is that the US is now a Banana Republic (I think Krugman warned about “turning Argentina” a few years ago), and we aren’t going to clean things up until we overthrow our oligarchy.

It strikes me as too simple, but I hope DeLong and/or Krugman take a whack at it.

The illusions of expertise

One the one hand, I don't buy the opinion of the expert Philip Tetlock. On the other hand, he would say I shouldn't (emphases mine) ...
Learning How to Think - Kristoff - NYTimes.com

... The expert on experts is Philip Tetlock, a professor at the University of California, Berkeley. His 2005 book, “Expert Political Judgment,” is based on two decades of tracking some 82,000 predictions by 284 experts. The experts’ forecasts were tracked both on the subjects of their specialties and on subjects that they knew little about.

The result? The predictions of experts were, on average, only a tiny bit better than random guesses — the equivalent of a chimpanzee throwing darts at a board.

“It made virtually no difference whether participants had doctorates, whether they were economists, political scientists, journalists or historians, whether they had policy experience or access to classified information, or whether they had logged many or few years of experience,” Mr. Tetlock wrote.

Indeed, the only consistent predictor was fame — and it was an inverse relationship. The more famous experts did worse than unknown ones. That had to do with a fault in the media. Talent bookers for television shows and reporters tended to call up experts who provided strong, coherent points of view, who saw things in blacks and whites....

Mr. Tetlock called experts such as these the “hedgehogs,” after a famous distinction by the late Sir Isaiah Berlin (my favorite philosopher) between hedgehogs and foxes. Hedgehogs tend to have a focused worldview, an ideological leaning, strong convictions; foxes are more cautious, more centrist, more likely to adjust their views, more pragmatic, more prone to self-doubt, more inclined to see complexity and nuance. And it turns out that while foxes don’t give great sound-bites, they are far more likely to get things right.

This was the distinction that mattered most among the forecasters, not whether they had expertise. Over all, the foxes did significantly better, both in areas they knew well and in areas they didn’t.
Sorry, I really do want to see a physician with expertise. I don't buy the extreme view that expertise doesn't matter.

I do, however, accept that fame is a reasonably reliable measure of incompetence. I'm a firm believer in the power of humility, which is probably a proxy for "self-doubt".

So I'd take a physician with both expertise and humility, and maybe more self-doubt than most would like. Spare me the famous physician. As a former scholar of evidence-based medicine, I can empirically confirm that the "opinions of the great" are frequently worthless.

The same thing, by the way, goes for CEOs. If they're famous enough to appear on a magazine cover, be suspicious.

Steve Jobs being an exception - but there's only one of him.

Ok. Bill Gates too.

Oh, wait, maybe I'm wrong?

But my self-doubt means you should trust me ...

The Ghost of CompuServe haunts the web

During a recent address book transfer I came across a CompuServe email address.

I'd forgotten about CompuServe. They were giants once. I had a CompuServe address in the 80s of course. I used their services on amber DOS terminals, with DOS and OS/2 character mode clients, and I think even with some form of Windows and Mac Classic GUI terminal.

That was around the type of Telenet and Tymnet and Fidonet and GENie and BIX and BBS services and the first version of Apple's many online services -- the one that became the basis for AOL (Mac only at first).

There's more, but it's a long, long time ago. Longer in experiences than mere years.

So I tried www.compuserve.com and I was redirect to - webcenters.netscape.compuserve.com/menu/default.jsp.

Yes, Netscape. Where the Search is "enhanced by Google" and there's an awkwardly formatted portal. The page is "Copyright © 2009 CompuServe Interactive Services, Inc". There's an "about" link (emphases mine) ...

CompuServe.com

CompuServe Interactive Services provides complete and comprehensive products and access for Internet online users at home, in the workplace and around the globe. With the launch of CompuServe 7.0 in 2001, CompuServe reached a new milestone by making the gathering of information and exploring the Internet faster, easier and more convenient than ever before for its worldwide membership. Since its acquisition by AOL in 1998, CompuServe has continued to enhance its core service to meet the needs of one of the fastest-growing segments of the Internet: value-driven adults going online for the first time.

An Internet Pioneer

Founded in 1969 as a computer time-sharing service, Columbus, Ohio-based CompuServe drove the initial emergence of the online service industry. In 1979, CompuServe became the first service to offer electronic mail capabilities and technical support to personal computer users. CompuServe broke new ground again in 1980 as the first online service to offer real-time chat online with its CB Simulator. By 1982, the company had formed its Network Services Division to provide wide-area networking capabilities to corporate clients.

CompuServe also led the interactive services industry overseas, entering the international arena in Japan in 1986 with Fujitsu and Nisso Iwai, developing a Japanese-language version of CompuServe called NIFTYSERVE. In 1989, the company expanded into Europe where it grew to be a leading Internet service provider.

A Key Brand

Since February 1998, CompuServe has been a wholly owned subsidiary of America Online, Inc. As part of the AOL Web Properties group, CompuServe plays an important role by providing Internet connectivity for value-minded consumers seeking both a dependable connection to the Internet and all the features and power of an online service.

The Ghost of CompuServe lives in the Ghost of Netscape within the still living corpse of AOL.

For a young world, the Net has a lot of ghosts.

Primates in zoos are not quite sane ...

Primates in zoos are mostly insane.

That's the lesson I take from Carnivorous gorillas | The Economist. It wasn't the point of the essay, but it's hard to escape the conclusion that they're not having much fun.

In a similar vein our local zoo has run through several psychotic Polar bears. They seem "ok" when they start out, but they show increasingly abnormal repetitive behaviors. They're going to get a much more impressive residence soon, so maybe the next batch will be merely neurotic.

Smart animals, shockingly, don't do well in confinement under continuous observations by history's most terrible large predator.

Let's not mention dolphins.

Wednesday, March 25, 2009

Google books is the best way to request library books

I wrote bout this two years ago, and again six months ago, but I think it’s worth a reminder. Google has a fantastic service for requesting books from my local library – and probably yours as well.

I’m not being selfless here. I’m worried Google is going to drop this lovely service; they’ve already removed the Google Profile “My Library” integration. Some more traffic might help.

It goes like this:

  1. Search in Google Books.
  2. Select your book (for example).
  3. Click “Find this book in a library”. That takes you to the WorldCat entry for the book. (What is WorldCat?)
    I don’t think you even have to create an  account on WorldCat [1]
  4. Enter your zip code (it remembers the last entry) and browse the list. It showed the book in several local libraries. I clicked on the Saint Paul Public Library link and went immediately the request page.

Unfortunately I still had to use the silly Saint Paul Public Library authentication procedure. That’s the one where one enters a longish ID number that’s masked during entry.

It’s faster and much more elegant than going directly against our library system.

[1] You can export WorldCat cites to reference management software. That takes me back.

Not the Great Recession

The name has been taken … repeatedly …

Let’s Call It a Flump - Freakonomics Blog - NYTimes.com

… the emerging favorite, the “Great Recession,” was struck down in a neat piece of research by Catherine Rampell, whose careful combing of the archives revealed that “Every recession of the last several decades has, at some point or another, received this special designation.”…

I suppose we shall have to defer the official name until 2012 or so …

Nokia never again - mobile phone unlocking lessons

I don't love Apple. I don't even love my iPhone, though I admit to occasional flirtations.

On the other hand, I really don't love the rest of the mobile industry.

My latest experience with unlove started with an unused AT&T issued Nokia 6555b. It's been sitting in the phone bin since I bought my iPhone, but recently I decided to use it with a Pay-As-You-Go account. It's to be a child phone -- especially for one son who tends to get lost.

Easy, I thought. It's GSM, so I'll just order the $10 T-Mobile kit and swap cards. I could have gone with AT&T's Pay-Go plans, but, I don't trust 'em and their plan looked typically squirrely.

Yes, for those of you of a certain age, that is the theme music from Jaws that you hear.

To cut to the chase, I've spent time with T-Mobile's automated phone activation system (very much unlove), T-Mobile customer support (surprisingly good) and AT&T's local store (saintly, really. Lousy company, but good staff).

Turns out Nokia has their own special phone locking procedure; they're supposed to make an unlock available for every phone that's been in use for more than 3 months.

Nowadays every Nokia phone is locked with a unique code, and supposedly only Nokia can provide it. The AT&T rep spent 30 minutes (this is the rep, not poor old me) working through the AT&T/Nokia system to get to the point where, in a week or two, they'll send me the unlock code. To a handwritten email address (and what's the chance that will work?).

Then, if I can find the original AT&T SIM card, and I can follow the bizarre series of incantations I was given, then maybe this Nokia phone, which I paid for, will be usable outside of AT&T.

I ain't buying Nokia again.

Apple, I unlove you less than the rest.

Update: Of course this isn't a new problem.

So why is it that today's NYT article on the virtues of unlocked phones overseas glosses over the varying policies of different phone manufacturers? Why doesn't Nokia get more flack for their obscure unlocking procedure -- even when AT&T is asking for the unlock?

And, for extra credit, how is this wee little episode deeply connected to the collapse of the global economic system?

Update 3/27/09: Astonishingly, it worked. The two reps, Nokia and AT&T alike, managed to get the email right and work through the process. Technical directions are on tech.kateva.org. AT&T and Nokia are lousy companies, but they do have some good employees.

Update 7/28/09: I get quite a few comment submissions for this post linking to unlocking sites. Many of these sites are vectors for injecting trojans into visiting computers. I won't allow any of those comments.

Olympian coding - a snapshot from iPhone game development

Daring Fireball recommends the story of John Carmack's iPhone Wolfenstein 3D development.

I second the recommendation. My jaw dropped when he casually mentions restarting with a DOS 286 code base. The attitudes towards code reuse, and the distinction between "commercial" and "modern" development are revealing.

A lot of respect for very old code that's proven good, but also a willingness to ship good-enough stuff.

Garrison Keillor's mother

Is Garrison pulling our leg?

The real American dream | Salon

...I took my mother fishing last year and discovered she'd been in the Johnson & Swanson Circus. She did backflips on a tightrope and swallowed flaming torches and exhaled a stream of flame 10 feet long. Recently we found a photograph of her in spangly tights, a hibiscus in her hair, standing blindfolded on the trunk of an elephant with a lit cigarette in her mouth which a swarthy man in a gypsy outfit is about to shoot out of her mouth with a pistol aimed over his left shoulder using a small mirror with a mother-of-pearl handle...

It would be like him.

It is also true, however, that even as a young doctor I knew some of those "spry" aged women with the proverbial twinkly eyes had stories they weren't about to tell me.

The problem with Afghanistan - we're broke

Kaplan has a good summary of the intelligent debate about how to approach Afghanistan (CT is counter-terrorism, this approach is said to be favored by Joe Biden) ...

Obama must choose this week between two radically different Afghanistan policies. - By Fred Kaplan - Slate Magazine

... Some in the CT camp realize that the COIN-dinistas (as critics call them) have a point. Their real gripe with counterinsurgency is that it costs too much and promises too little. Even most COIN strategists acknowledge that a successful campaign, especially in Afghanistan, would require lots of troops (way more than President Obama has committed so far), lots of time (a decade or so), and lots of money (wiping out most or all of the savings achieved by the withdrawal from Iraq)—and even then the insurgents might still win.

A "targeted" CT campaign, its advocates say, would at least demonstrate the West's resolve in the war on terrorism and keep al-Qaida jihadists contained. It's a type of fighting that we know how to do, and its effects are measurable. One might also argue (I don't know if anyone on the inside is doing so) that it could serve as a holding action—a way of keeping Afghanistan from plunging deeper into chaos—while we focus more intently on diplomatic measures to stabilize neighboring Pakistan. If Pakistan blows up, curing Afghanistan of its problems will be irrelevant and, in any case, impossible.

Some in the COIN camp have sympathy for this argument—especially for the part about the high cost and the uncertainty of success—but they would argue back that a purely CT approach is sure to fail in the long run...

I think we ended up going the COIN route in Iraq. It's too early to know how well it worked, but it seems to have been an improvement on than every other approach that was tried there.

One thing we know, however, is that we're financially and militarily exhausted.

So the real debate is between containment, and a Pakistan stabilization strategy, and a big investment in Afghanistan.

I sympathize with everyone, and especially the people of Afghanistan.

The one consolation is that I think we have a good team struggling with the problem.