Friday, January 26, 2007

Higgs?

A particle physicist dares to speculate that his team has spotted the Higgs boson. Odds are this is a false alarm, but there's a decent chance its real. Readers of Cosmic Variance have a ringside seat. Either we'll see the pain of everyday science or the joy of a momentous discovery.

The comments are quite good.

How to hack a human: start with the insula

Say you want to hack a human. You want to alter what they love, what they hate, what they want. You probably start with the insula:

In Clue to Addiction, Brain Injury Halts Smoking - New York Times

... The patients’ desire to eat, by contrast, was intact. This suggests, the authors wrote, that the insula is critical for behaviors whose bodily effects become pleasurable because they are learned, like cigarette smoking.

The insula, for years a wallflower of brain anatomy, has emerged as a region of interest based in part on recent work by Dr. Antonio Damasio, a neurologist and director of the Brain and Creativity Institute. The insula has widely distributed connections, both in the thinking cortex above, and down below in subcortical areas, like the brain stem, that maintain heart rate, blood pressure and body temperature, the body’s primal survival systems.

Based on his studies and others’, Dr. Damasio argues that the insula, in effect, maps these signals from the body’s physical plant, and integrates them so the conscious brain can interpret them as a coherent emotion.

The system works from the bottom up. First, the body senses cues in the outside world, and responds. The heart rate might elevate at the sight of a stranger’s angry face, for example; other muscles might relax in response to a pleasant whiff of smoke.

All of this happens instantaneously and unconsciously, Dr. Damasio said — until the insula integrates the information and makes it readable to the conscious regions of the brain.

“In a sense it’s not surprising that the insula is an important part of this circuit maintaining addiction, because we realized some years ago that it was going to be a critical platform for emotions,” Dr. Damasio said in a telephone interview. “It is on this platform that we first anticipate pain and pleasure, not just smoking but eating chocolate, drinking a glass of wine, all of it.”

This explains why cravings are so physical, and so hard to shake, he said: they have taken hold in the visceral reaches of the body well before they are even conscious. ...

Between the cortex and the subcortex, a processor that translates sensations into emotions, wants, feelings. Humans will do bad and good things with this.

Thursday, January 25, 2007

The NeXT Years: Steve Job

Holy cow. The NeXT Years: Steve Jobs is not exactly the CEO story one reads in the Harvard Business Review. It's a raw mess of chaos, brilliance, randomness and mass delusion that somehow produced a vast amount of wealth -- for someone. Even Canon might have got a bit of their NeXT investment back.

The article is all about Jobs, who is both appalling and fascinating in roughly equal measures. It's obvious there are some other very important minds that are doing the real work under the radar, but their stories are probably less scandalous.

Despite himself, Jobs ends up being inspirational. He was despised, discarded and abandoned, but he kept coming back. It's a story worth remembering when misfortune strikes; it's truer and more useful than the usual fraudulent tales of CEO perfection.

I wonder what Jobs parents make of him ...

All DeLong all the time: Egregious Moderation

Can someone tie Brad DeLong down before he exhausts the rest of us? In addition to his personal blog and shrillblog, he's now launched ...
Egregious Moderation

...An egregiously moderate forum: for people who want one online source for punchy liberal analysis and evisceration; especially evisceration...
He reads voraciously, writes incessantly, and is widely believed to be a full-time professor and productive economist. One theory is that he's an early experiment in shared-consciousness clonal breeding.

I need to create a "DeLong" category in my bloglines feed...

Wednesday, January 24, 2007

Why I didn't renew my Harvard Business Review subscription

HBR did one of their typical slavish portraits of Robert Nardelli a few months back...
101 Dumbest Moments in Business | 41 | Business 2.0:

... Dodging investors angry over the pay received by Home Depot chairman and CEO Robert Nardelli, who took home at least $120 million over five years as the company's stock price dropped 12 percent, Home Depot's board fails to show up at its annual shareholders meeting.

The session is presided over solely by Nardelli, who sidesteps all questions ('This is not the forum in which we would address your comment') and cuts the meeting short after half an hour. The event's negative fallout, highlighted by demonstrators wearing chicken costumes and orange Home Depot aprons, leads Nardelli to announce days later that, for next year's meeting, 'we will return to our traditional format ... with the board of directors in attendance.'

Nardelli resigns in early January, walking away with another $210 million in severance.
I dropped my HBR subscription last fall. HBR is the opiate of the powerful.

BTW. The CNN 101 Dumbest Moments in Business is quite good.

Even Robert Reich had something nice to say about Bush

I'm a Robert Reich fan. I know Reich is no more likely than I to say nice things about Bush. Like me, however, he was impressed by one Bush proposal...
Robert Reich's Blog

...the only halfway interesting thing about the President's underwhelmingly platitudinous State of the Union speech was his health care proposal. It deserves one cheer for the following reason: It potentially de-couples health care from employment.

Under his proposal, everyone would be eligible for a tax deduction for health insurance up to $15,000 per family, $7,500 for a single person – regardless of whether the insurance is provided by the employer or purchased elsewhere. And there would no longer be any advantage to getting it at work because employer-paid premiums would be included in taxable income.

Get it? With this plan, you can just about kiss employer-provided health insurance good-bye.
And good riddance. It’s the biggest tax break in the whole federal tax system, costing the Treasury some $130 billion a year. But you’re not eligible for it when you and your family are most likely to need it – when you lose your job, for example. And the biggest beneficiaries are upper-income employees. The lower your pay, the less likely you are to get any employer coverage at all.

The current employer-based system doesn’t cover the self-employed – the largest and fastest-growing category of worker. And it creates perverse incentives. It encourages employers to seek out young, healthy employees who are unlikely to have health problems; reject older ones; and push married employees onto their spouse’s employer’s plans...
I still think Bush and his minions would screw this up if they ever really tried to implement it, but I'm with Reich. Any healthcare reform plan that doesn't separate employment from health insurance should be abandoned.

Microwave sponges: two minutes on full

Health concerns aside, you can save money on sponges ...
Microwaves turn kitchen cloths into germ killers | the Daily Mail

...Two minutes of microwaving on full power mode killed or inactivated more than 99 per cent of all the living pathogens in the sponges and pads.

The Bacillus cereus spores required four minutes for total inactivation.

Professor Bitton said the heat, rather than the microwave radiation, was the most likely cause of death for the pathogens. As the microwave works by exciting water molecules, it is better to put wet rather than dry sponges or scrub pads into the oven.

... Cooks should microwave their sponges every other day, he suggested.

The warm, damp environment of kitchen cloths is the ideal breeding ground for microbes.

In the right conditions one bacterium can multiply to more than four million in just eight hours. This can make them up to 200 times more infested than a lavatory seat.

Monday, January 22, 2007

Spam: state of the art report

MSNBC's Rob Sullivan has a spam report. The numbers are indeed staggering. I wonder what percentage of net traffic is made up of "high grade" material -- excluding spam, porn, illegal file sharing etc. I'm guessing it's in the 20-30% range overall. A surprising amount of net traffic now is file sharing, and it's widely believed that almost all of that (by volume) is copyrighted material. Emphases mine.

... Not long ago, there seemed hope that spam had passed its prime. Just last December, the Federal Trade Commission published an optimistic state-of-spam report, citing research indicating spam had leveled off or even dropped during the previous year.

Instead, it now appears spammers had simply gone back to the drawing board. There's more spam now than ever before.

In fact, there's twice as much spam now as opposed to this time last year... About half of all spam sent now is "image spam," containing server-clogging pictures that are up to 10 times the size of traditional text spam. And most image spam is stock-related, pump-and-dump scams which can harm investors who don't even use e-mail. About one-third of all spam is stock spam now.

... There are 62 billion spam messages sent every day, IronPort says, up from 31 billion last year. Now, spam accounts for three of every four e-mails sent, according to another anti-spam firm, MessageLabs.

Image spam is a big part of the resurgence of unwanted e-mail. By using pictures instead of words in their messages, spammers are able to evade filters designed to detect traditional text-based ads. New computer viruses have contributed to the uptick, also, particularly a surprisingly prolific Trojan horse program called "SpamThru" that turns home computers into spam-churning "bots."

... Stock spam is effective because no Web link is required, Cluley said. In old-fashioned spam, criminals generally try to trick recipients into clicking on a link and buying something. Many e-mail programs now block direct Web links from e-mails, rendering click-dependent spam much less effective. But stock messages merely have to make the recipient curious enough about a company to motivate him or her to buy a few shares through a broker.

There is another element that helps perpetuate stock spam, Stark said – he believes speculators unrelated to the original spam sometimes try to “play the momentum” surrounding a spam campaign – either getting in early on a pump-and-dump campaign to profit as shares rise, or by “shorting” stocks, betting that they will fall after the spam campaign flames out.

...

Image spam, which seems not inseparable from stock spam, can arrive entirely devoid of text, but that’s not common. Most messages have what appears to be nonsense text pasted above and below the image. Experts call this "word salad," or "good word poisoning."..

... The word jumble is generally borrowed from news headlines or classic books like Charles Dickens' “David Copperfield,” the text of which are often available online. The seemingly random text actually serves and important purpose -- to foil or confuse word-based spam filtering.

... Spammers continually refine and combine their techniques, said Doug Bowers, senior director of anti-abuse engineering at Symantec. The firm recently found spam attached to legitimate newsletters that appear to be from big companies, including a Viagra ad atop a 1-800-Flowers e-mail newsletter and another on an NFL fantasy league letter. Such e-mails are simply spam masquerading as authentic, with real content borrowed from legitimate companies. They are similar to phishing e-mails, and so are much more likely to be opened by recipients than traditional spam, Bower said...

Natural selection is causing spam to evolve very quickly. We're recreating biological evolution at a frenetic pace. Defense requires more complex algorithms, which lead quickly to more complex attacks. Maybe every technological civilization succumbs when its spam becomes sentient ...

The stock tip churn process may work for quite a while. It will eventually become a contest between spammers and speculators, which each speculator hoping they can hop off fast enough before the "house" calls the game. Of course the spammers will always know more, so they'll always come out ahead. Some speculators will win too, so it will be a lot like going to the casino. In time the spammers will learn to keep the game interesting.

My favorite spam fighting technique, the reputation management of authenticated sending services, works even against spambots. I think this is what Google is doing now, even though they're very quiet about it.

Sunday, January 21, 2007

John Edwards' blog

Anyone would be better than Bush, and any of the current Democrats running for president would be better than McCain. Of them all, however, I have a soft spot for Edwards. He seems the closest to Al Gore, who, after all, would have won cleanly if not for Nader.

The media, currently infatuated with Obama (smokes cigarettes?! I can't get my head past that), doesn't mention Edwards at all. Unsurprisingly, he has a blog. I'm going to read it for a while, and pass on anything interesting he might say. It's a change from Hilary/Obama anyway ...

Humans: 70,000 years old

When did humanity start? One marker is the time we upgraded the 1 million year old stone axe. DeLong excerpts Lapite who quotes Hawks:
Grasping Reality with Both Hands: Brad DeLong's Semi-Daily Journal: The Dawn of Humanity

[Lapite] ... Now, the question you have to ask yourself is just how "human" creatures incapable of bettering the simple stone handaxe over a million years could possibly be; they may have looked like us, but it's clear they didn't think like us, and the timespans under consideration rule out "culture" as the limitation here.

Indeed, as Hawks suggests, at this point it isn't even clear that such a thing as "culture" (and its attendant variation across time and space) existed in a meaningful sense until about 80,000 years ago...

There was a big wetware upgrade 80K years ago. Hawks and others have pointed out other likely upgrades 40K and maybe 15K years ago. The process is continuing; natural selection doesn't stop just because we our world is increasingly artificial and virtual.

I read Hawks and DeLong, therefore I shall sample Lapite for a time ...

Saturday, January 20, 2007

Bush: something good?!!

Greg Mankiw's notes that Bush has a healthcare proposal
that moves us away from employer-purchased health insurance. I reply:
Hell has frozen. Someone is addressing the pathologic tie between employers and health insurance, and it's a man renowned for his incompetence and malign leadership -- GW Bush.

I'm certain whatever he really proposes will be a monstrous screw-up or a feint to cover for some malign measure, but the summary being circulated is right and good.
I know Bush will screw this up. There's a chance though, that with a democratic congress something good may come of this.

Any fix for American healthcare will displease a large portion of the middle-class. So any fundamental fix is political poison. It won't happen.

Fixing the employment-healthcare bond, however, is politically feasible. It would be a great leap forward.

1/25/07: I did say I knew Bush would screw this up.

Cultures We Really Evolved that are Stranger Than Any You can Think of

John Crowley Little and Big - The List is a non-fiction reading list of stranger than strange humanity. It was written in part to provide fodder for would be fantasy writers. Fascinating. Thanks Neil Gaiman for the pointer.

Thursday, January 18, 2007

We don't have Exxon to kick around any more ...

Crooked Timber reports Exxon has given up on the "no global warming" project. New CEO apparently. They've dumped their puppet pressure groups and will now focus on minimizing the impact on Exxon of remediation efforts.

The WSJ must be feeling very lonely.

There is real ongoing debate however -- about how much to do, how much to spend, what measures to take, who pays the price, etc.

I'll miss Exxon. They were evil in a particularly clumsy way.

The answer to the Fermi Paradox? Alas, probably not.

Personally, I think an answer to the Fermi Paradox belongs in Science, not New Scientist:

So much space, so little time: why aliens haven't found us yet | Science | Guardian Unlimited

... It ranks among the most enduring mysteries of the cosmos. Physicists call it the Fermi paradox after the Italian Nobel laureate Enrico Fermi, who, in 1950, pointed out the glaring conflict between predictions that life was elsewhere in the universe - and the conspicuous lack of aliens who have come to visit.

Now a Danish researcher believes he may have solved the paradox. Extra-terrestrials have yet to find us because they haven't had enough time to look.

Using a computer simulation of our own galaxy, the Milky Way, Rasmus Bjork, a physicist at the Niels Bohr institute in Copenhagen, proposed that a single civilisation might build eight intergalactic probes and launch them on missions to search for life. Once on their way each probe would send out eight more mini-probes, which would head for the nearest stars and look for habitable planets.

Mr Bjork confined the probes to search only solar systems in what is called the "galactic habitable zone" of the Milky Way, where solar systems are close enough to the centre to have the right elements necessary to form rocky, life-sustaining planets, but are far enough out to avoid being struck by asteroids, seared by stars or frazzled by bursts of radiation.

He found that even if the alien ships could hurtle through space at a tenth of the speed of light, or 30,000km a second, - Nasa's current Cassini mission to Saturn is plodding along at 32km a second - it would take 10bn years, roughly half the age of the universe, to explore just 4% of the galaxy. His study is reported in New Scientist today.

Like humans, alien civilisations could shorten the time to find extra-terrestrials by picking up television and radio broadcasts that might leak from colonised planets. "Even then, unless they can develop an exotic form of transport that gets them across the galaxy in two weeks it's still going to take millions of years to find us," said Mr Bjork. "There are so many stars in the galaxy that probably life could exist elsewhere, but will we ever get in contact with them? Not in our lifetime," he added. ...

He gets a different answer from Fermi because there’s no exponential growth. The 8 probes have 8 subprobes, so it’s only 64 probes exploring the galaxy. Fermi assumed the travel of sentient organisms that then launched new exploration from each “residence”, so there were long residence times but there was exponential “probe” growth. Exponential growth conquers all.

Most modern formulations of the paradox assume self-replication of abiologic entities — the probes have the capability to construct copies of each other, so in a few generations there are trillions of probes.

If we find there’s some immense obstacle to self-replication then this result stands. If not the paradox stands. The most common resolution to the paradox is that technologic civilizations do not sustain an interest in travel, so Bjork’s answer is (to me) very enticing — it suggests a more optimistic answer. I’d describe Bjorks’ answer as “self-replicating machines cannot be created within the lifetime of technological civilizations”. So the difficulty of creation would set an upper bound on the lifetime.

The Guardian is misleading about where he published, btw. He published in an online repository, New Scientist just reported on it. This is interesting enough to merit updating my Fermi Paradox page however.

PS. This Slashdot comment is hilarious. The low rating is why I don't usually read Slashdot comments -- the raters are dysfunctional. (Comments are often good, it's the ratings system that's broken)

.. Negative. I find your argument untenable. I am in agreement with the Danish monkey-being. Probabilities of non-human life spreading through the Galaxy and discovering primitive monkey-beings in Sol System are minimal. Probability is on the same order of probability of a F'narthag slime-weasel evolving wings and taking flight. It is also highly improbable that extraterrestrial beings would colonize the pathetic planet Earth and blend into the primitive monkey-being society. They would be forced to hide in internet discussion groups and the tech sector so that they are mistaken for geeks when they display lack of monkey-being social skills....

The Slashdot commentators had the same response I did -- ignoring self-replication is nonsensical. The author came up with a silly reason why self-replication wouldn't work; he said the probes interfere with one another. Duh.

Sprint and Motrola conspire to sell me an AT&T (Cingular) iPhone

Sprint and Motorola must be conspiring to raise the share prices of AT&T (once Cingular) and Apple. I think the SEC should investigate.

How else can I explain the Motorola RAZR V3M? I bought this phone from pure necessity — an old phone had died and I needed a replacement. I don’t like any phone on the market today; this phone seemed to the best choice of a bad lot. A mixed bag of features, but the list price (which I paid, no contract extensions!) was about $200 or so.

Turns out, the phone is better than I’d expected. It took some significant effort, and lot of help from Google, but it’s possible to beat the Sprint out of it and make it a half-useful device with only a handful of atrocious usability flaws. True, Sprint’s high-speed network sucks the battery dry, but that wouldn’t be so bad if it weren’t so difficult to recharge it. Alas, that’s where the worm turns.

There are two big problems with the phone. One is the battery drain with data use, compounded with a difficulty in recharging. Despite the min-USB cable connector, it won’t charge from a conventional USB charger or a car USB power outlet. It will charge from a computer USB port — but only if one installs the (semi-secret) XP-only modem enabling utility. The other problem is that it can’t be used to attach a computer to Sprint’s data network — unless one installs the XP-only modem enabling utility.

The inability to accept a charge from a standard USB charger is highly annoying (it should), but I could live with that that if the RAZR would charge from my MacBook, iBook, iMac, etc. I could especially live with it (albeit gritting my teeth) if the phone would put my MacBook on the net when I’m not on a WiFi connection.

Alas, it’s not to be. Sprint and Motorola are simply insisting that I move to Apple/AT&T. Did Apple secretly give the CEOs of Sprint and Motorola backdated Apple stock options?