Dean's World

Chuck Simmins noted problems with World Health Organization numbers on SARS. And problems with their numbers on the marburg outbreak in Angola. And with their latest wild numbers on Bird Flu. I won't even get into their numbers on HIV.

A first response on something like this would be, "hey, they're doing the best with what they have." But are they? How much of what they do is really politics, as opposed to sound science?

Oh cool... here are some free public lectures by Stephen Hawking.

As is usually the case with Hawking, his stuff requires careful reading and re-reading and pondering at length 'lest he be misunderstood. Hawking often surprises me and baffles me until I read him more closely and really think about what he's saying. I always feel immeasurably smarter after I feel I've grasped Hawking (which, come to think of it, is a blindingly obvious sentiment).

For example, in the main lecture linked above, he makes the following observation on the Anthropic Principle:

For the Strong Anthropic Principle, one supposes that there are many different universes, each with different values of the physical constants.

I found myself utterly baffled by this at first, since there's nothing in the Anthropic Principle, weak or strong, which states such a thing. Just for the record, here's how I understand the strong and weak anthropic principles:

Strong anthropic principle: the universe is ordered in such a way that the development of intelligent life is inevitable, in much the same way that the development of molecules or stars are inevitable. It simply will happen. Thus the laws of the cosmos must (by deduction) be ordered in such a way that the odds of intelligent life developing sooner or later approach, or are equal to, 1.

Weak anthropic principle the universe is ordered in such a way that intelligent life is something that might possibly happen, with no great likelihood. Odds of sentience developing in the universe approach, but do not reach, 0. (They can't equal 0 or we wouldn't be here to observe ourselves.)

That's how I understand it anyway; anyone who knows better can correct me. However, I find these head-scratchers a lot in cosmology, especially reading guys like Hawking or Einstein, because they're often several jumps ahead of their audience--and certainly several jumps ahead of me. As I grasp it, Hawking isn't mis-stating the Strong Anthropic Principle, he's stating that if the Strong Anthropic Principle is true it probably requires many other universes to exist in which life doesn't or can't exist, and he finds that unsatisfying. Thus he prefers the Weak principle. Which is fascinating, and I'm re-reading the essay now to make sure I understand what leads him to that level of assumption.

That's how I'm reading it anyway. Maybe someone will contradict me and tell me if I'm off-base...?

Anyway, look at this amazing treasure: FREE STEPHEN HAWKING LECTURES! Oh the wonders of the Internet. Thank you, Al Gore. (And you know, some time I'll have to write about why that's not as big a joke as some people think it is.)

This has to be the most hilarious picture I've seen all year. (Remember, it's still January.) Pictured at right is Canadian Sasquatology expert Thomas Steenburg. He and his fellow pioneering researchers made that cast themselves, they did.

Honest to God, could you even imagine making a more transparently fake-looking footprint than that? Oh, I pine for the good old days, when the real Sasquatch researchers would put some effort into their fakeries. I mean, look at those toes!

I want that thing for an ashtray.

I stole the photo shamelessly from this Globe & Mail story about the cutting edge of Bigfoot research, by the way. Read the whole thing. The best line: "A lot of the old guys had set ideas about what a sasquatch is and what it does. We've got some new ideas. . ."

Okay, I may be wrong. It is true there is at least one credentialed wildlife biologist who thinks sasquatches are probably real. And it's not like there's no place in the world for researching oddball things like this. But if a half century goes by and no one can come up with a single solid photo or film, not one skeleton, not one hair sample, not one accidental shooting, not one thing besides footprints and uncertain eyewitness accountings for a massive species that supposedly roams from Northern California all the way up through British Columbia? It's hard not to laugh.

Yeah yeah, the laugh will be on us if they find one of these things. I'll be the first to admit I was wrong and to apologize for laughing, in fact. But I ain't holding my breath.

Anyway, it's a harmless hobby either way, eh?

(Thanks Robert.)

Quoted:

Nearly every chart told the same story. Boys are over 50 percent more likely than girls to repeat grades in elementary school, one-third more likely to drop out of high school, and twice as likely to be identified with a learning disability. The response? Near-total silence.

What's most worrisome are not long-standing gender differences but recent plunges in boys' relative performance. Between 1992 and 2002, the gap by which high school girls outperformed boys on tests in both reading and writing-- especially writing-- widened significantly. Given the reading and writing demands of today's college curriculum, that means a lot of boys out there are falling well short of being considered "college material." Which is why women now significantly outnumber men on college campuses, a phenomenon familiar enough to any sorority sister seeking a date to the next formal. This June, nearly six out of ten bachelor's degrees awarded will go to women. If the Department of Education's report is any indication, in coming years, this gender gap will grow even larger.

The report illustrates a dramatic and unsolved mystery: At some point in the early '80s, boys' relative academic records and aspirations took a downward turn. So far, no one has come up with a good explanation for this trend, but it's a story that affects millions of boys and their families. And yet, according to LexisNexis, the report was cited by name in only five newspaper and magazine articles.

More at The New Republic, which is worth registering for, and not just for this one article, either.

Via Professor Reynolds, who has more links.

I'm reading a fascinating book about programmable electrons and artificial atoms. It's called Hacking Matter and it's really quite remarkable. It's available in book stores, or you can download it from author Will McCarthy's web site.

I must say it's a mindblower. Artificial atoms? Programmable electrons and photons? And not just in theory, but actually demonstrated in a lab, and in a few cases available for sale from companies specializing in these technologies?

The future looks stranger every day.

I see that they're getting even more advanced in using inkjet-style printer technology with living cells.

Neat to think that one day we may be able to fix people with horrible abrasions or burns by just spraying on new skin, isn't it? Or fix other things that way...

Okay, I had to look and see if it was April Fool's Day, but apparently it isn't: Taiwanese scientists claim to have put jellyfish genes into pig embryos to produce green, glow-in-the-dark pigs.

The pigs are green-tinged in daylight, and glow in the dark at night.

Well. That's something we've always needed.

(I also assume Kermit and Miss Piggy were somehow involved but are just too shy to make a public statement.)

William Saletan mentions that studies have shown that “In teens, a sleep-inducing hormone doesn't start rising till 10 or 11 p.m. and doesn't let up till 8 a.m.” in other words Teens aren’t actually very awake until about 9am or so. Yet schools tend to start around 7:30am...

The best worst bit was this “Skeptical parents say adjusting the school day would 1) interfere with after-school jobs [...]" In case anyone missed it some parents would rather have their kids working than learning. This must be why our test scores are so low...

Quote:

"With the exception of a single cloned dog, all the major scientific discoveries claimed by Woo Suk Hwang, once the world's leading stem cell scientist, were faked, a university panel concluded today."

Whoops.

More right here.

A recent discussion brought this article to mind again, so I thought I'd link it again. It's about a troubling problem in science: when you only accept one hypothesis as valid to explain certain phenomena, that hypothesis tends to lose value, for it ceases to make predictions. Rather, you start finding ways to fit every observed phenomenon into the hypothesis--since you're only accepting one, you wind up having little choice but to do so.

He makes the point better than I do. Here it is again: The Problem With Single Hypotheses.

It applies somewhat to everyday life too, come to think of it. A lesson I probably ought to remember more often myself.

Update: Whispers in the airstreams claims that single hypotheses are only a problem in politics, not science. To be honest, it's hard for me not to snort derisively at that (no offense intended).

First, I snort because anyone who thinks that the current administration is the only to meddle in science is simply ignorant of history: there has never been an administration that hasn't done this, at least not since government began funding science.

Second, since the government funds such huge amounts of science today, and most of the rest is funded by major corporations, most research is now political by default. And it should be. You do not get to demand millions or billions in taxpayer money, and then declare yourself above or immune to politics. Furthermore, the competition for the grants is often entirely political, and dissident scientists are often shut out simply because they aren't in the majority--therefore they get no grants, therefore their careers stall or end.

Finally, Whispers claims "Science has a means to assure that intellectual integrity will be the ultimate outcome, politics doesn’t." And what means would that be? The peer review process? The same one which has time and again been shown to make major mistakes, and to embrace failed paradigms for years, decades, even generations? Take a good look at outcomes in politics and I suggest you won't find much different. If politics is war by other means, then peer review is politics by other means.

Jewett is right: in science, any science, when you allow only one hypothesis to be entertained, you cease to be practicing science. You're merely practicing belief, and fitting whatever facts you find into that preconceived belief. You need a second hypothesis, even one you don't believe in, as a foil to test your assumptions against.

This is interesting: a gene called P11 and its related proteins may play a role in chronic depression. Those of us who have problems with this particular malady are always on the lookout for stuff like that.

There's an old question when it comes to the brain: if you see differences in brain construction associated with a behavior or malady, does that mean the oddity you see caused the problem, or did the brain difference you see simply come about due to the problem? It's not a trick question. For example, pretend you know nothing about alcoholism, but you discover that most alcoholics post-mortem have enlarged livers. So did the alcoholism cause the enlarged liver, or did the enlarged liver cause the alcoholism? Well today we know (or are pretty sure), but go back far enough, to when they just started studying such questions, and you wouldn't be so certain.

If you look at the brains of chronically depressed people, you find certain chemical or structural differences. But are those differences the cause of their depression? Or is it that being depressed all the time causes the differences you see? The distinction is crucial; imagine trying to cure alcoholism by giving someone liver-reduction pills and you'll see the hazard.

We know for a fact that people's brain structures shift throughout their lives, not just as a part of aging but also as a part of activity; if you take up violin lessons late in life, for example, even if you've never taken music lessons before, certain parts of your brain will develop in ways that scientists can measure--at least if you keep up with your lessons. The brain literally changes as you change, depending on what you do with your life. When it comes to something like, say, depression, anxiety, schizophrenia, etc., little discoveries like this help get us closer to a stronger understanding of the exact nature of the problem.

I'm continually amused by how many people I encounter who think Ray Kurzweil's concept of the approaching human singularity as some sort of Rapture for Nerds. This is ridiculous, and is almst entirely coming from people who haven't read the book and have decided they know what's in it and what it says and that it's a bunch of nonsense.

How do you even have a discussion with such people? What it amounts to is, they don't like the word. Plus, they want very badly to dismiss the ideas behind it.

But consider: it took over a decade to run the sequence for one of the simplest viruses on the planet, HIV, which is only nine genes long, two short chromosomes. When the Human Genome project started, it was greeted with skepticism; yet the dozens of chromosomes and tens of thousands of genes took less time to sequence than HIV did, and was completed ahead of schedule. Now we routinely run genome mappings in periods measured in months.

That's the power of exponential growth, and that's what's at the core of Kurzweil's analysis. These exponential growth curves in computing technology, artificial intelligence software, nanotechnology, and biotechnology are going to completely change what it means to be human in a very short time period. Consider the following statements that were all valid not long ago:

Only a human can drive a car.
Only a human can master the game of chess.
Only humans can recognize an individual face.
Only humans can understand continuous speech.
Only a human musician can compose music in the style of Bach.
Only a human can improvise jazz.
Only a human can pilot an airplane.

Those were all true 10 years ago. They're all false now. We went from having the "holy grail" of robotics being the ability to walk upright on their own just ten years ago, to having such robots not only walking, but jumping, dancing, balancing on one foot... and doing that while talking, following spoken commands, recognizing faces, able to get back up if you trip them, and run simple office errands--such robots will be on sale by the end of this year.

We can now create artificial ears that fix most forms of deafness by wiring them straight into the nervous system. We're experimenting with chip implants to go into the brains of Parkinson's patients to compensate for their loss of motor function. We've got nanomachines working on their own assembly lines, self-regulating, and nanomaterials that are several times lighter and stronger than steel. And all of these technologies and more are on exponential growth trends, with the most pessimistic estimates still showing no sign of major slowdowns for the next two decades.

We've got experiments working right now where monkeys can remotely control robot arms based entirely on wires we've embedded in their brains. We could do it in humans now but we need the clearance to do it. Why do it? To help paraplegics of course. But once we can do that as routinely as we now give implants to fix deafness, we're going to be able to start thinking about other ways to enhance the brain.

People who suggest that it's all fleeting and going to come crashing to a halt remind me of the people that Gregg Easterbrook identified in The Progress Paradox: How Life Gets Better While People Feel Worse. By any measurable--I said measurable--standard, the human condition is improving in the vast majority of the world. You name it--health, lifespan, clean water, clean air, abundant food, leisure time, health, safety, security, mobility, education, it's all getting better. Yet as this happens, people are convinced that things are getting worse anyway.

Methinks most of Kurzweil's critics are guilty of a similar type of thinking: we can't possibly be on such an amazing technological cusp can we? Surely something must make it all come crashing to a halt soon, right?

I'm still reading Kurzweil's The Singularity Is Near: When Humans Transcend Biology. I'm about halfway through, having just finished the section on reverse-engineering the human brain and am now deep into his analysis on recent developments in nanotechnology and what the near-term and somewhat long-term future hold. I'm fairly convinced he knows what he's talking about and is right; barring a genuinely catastrophic series of events, human beings are going to be looking at a new level of evolutionary developent by the middle of this century at the latest.

I do wonder about the near-term future though, especially with some of the games we're playing with genetic manipulation and nanotechnology in particular. To explain what I mean, I note this device. I wonder, do any Dean's World readers recognize this? Some of you are old enough to have seen one and maybe even used one:

You can click on it to blow it up. Know what it is?

This is a device that was once popular in shoe stores. That's right, shoe stores. It was used usually to show parents how well their kids shoes fit.

See, the way it worked was, an tube in the box would be turned on, and would bathe the shoes and feet in x-rays, projecting a live image of the feet as they sat in their shoes. Three little viewing holes at the top would allow friends, family members, and store employees to see live x-ray images of your feet as you wiggled them in the shoes so they could see how well they fit.

This was a good example of a cheap technology that got cheaper and more readily available all the time, just as most technologies do. It was also hideously irresponsible; unlike modern fluoroscopes, which generally give you extremely short bursts of tightly controlled x-rays, this was live footage, a constant bombardment of a barely-shielded x-ray tube. I have no idea how many Grays of radiation were absorbed in an average shoe-fitting session, but I imagine most modern radiologists would shudder. You also have to wonder how well-shielded those devices were and what the long-term consequences were to employees who worked with and around those machines all day.

Take a look at this article on the boxes, and contemplate the interesting fact that these machines first came into use in 1940 in shoe stores, and were still in use in some states until the 1970s.

I don't know that we can be expected to conceive of every possible danger of the new technologies we're exploring, or stop in advance any mistakes we're going to make. But we are going to make them. They're putting nanoparticles on fabrics and in things like suntan lotion these days. Some researchers are right now currently working on building the first fully artificial lifeform, which they expect to have pretty soon. Meanwhile, machinery to let you analyze, play with, and modify viruses are now available in the $10,000 range and will only get cheaper all the time.

Kurzweil is right I think that most people think technological growth is linear. In the last hundred years, it hasn't been, it's been quite exponential, doubling every few years--in some cases, doubling annually or even faster. We're barelling ahead with a lot of these things while barely knowing what we're doing.

I tend to be optimistic about humans and human nature. I think we can overcome most threats we face. But there's no doubt we're going to make some mistakes. Some of them may be big ones.

I was just chatting this morning with my friend Matthew Stinson, who's currently living in China teaching English. By the way, how cool is that? I now routinely have instant message chats with people in Japan, China, the UK, and Afghanistan. And it's no big deal. Anyway, I thought this exchange might foster some discusstion:

[07:56] mattjs76: Just read your Atheist vs. Anti-Religion post. I had a moderately related conversation in class on my last week of the term.
[07:57] mattjs76: A student wouldn't let go of the religion topic even though I said it wasn't good for us to discuss. It's officially frowned upon by the higher-ups. Anyways...
[07:58] mattjs76: He goes to me, "The West is very developed, but are there really people there who believe God lives in the sky?"
[07:58] mattjs76: I replied, "You can be developed if you believe there's no God and you can be developed if you believe there is a God in Heaven. It's when you believe that God lives on earth that you run into trouble."
[07:59] EsmayDean: Good line.
[07:59] mattjs76: Historically, the Power Atheists have mocked the idea of a God in Heaven while saluting the idea of a God that lives on earth.
[08:01] mattjs76: I had another student confidentially email to me, and he lamented the way that science had created a new fatalism that mirrored the worst of religious fatalism.
[08:03] EsmayDean: Mind if I post this snippet of conversation?
[08:04] mattjs76: Not at all.
[08:06] mattjs76: The student is a math major. This is what he asked me, but I haven't yet replied to him: "I do not believe fatalism, and I always hold in esteem for science research, logic-thinking and suspect-thinking. So I devote myself into finding the basic result of most phenomenon. I discover that (maybe not only), each process in the world can be discribed by one or more equations. If we grasp all the factors which effect this process, we can foretell the future from a very precise equation.
How awful! That's to say after we struggle out of one "fatalism", we sink into another! Science, when it is highly advanced, we can see it is so similar with religion. But if science is replaced by another vanguard subject (just as religion is replaced by science), what's our imagination of the next thing?"
[08:07] EsmayDean: Yes.
[08:07] mattjs76: I think we already saw an attempt to rebel against science with the Postmodernists. They're weak now, thankfully, but they can always come back.
[08:07] EsmayDean: I've known some so-called scientists with degrees but no apparent ability to think independently to say the most mind-bogglingly stupid things.
[08:08] EsmayDean: Amazingly intolerant too, some of them.
[08:08] EsmayDean: Most aren't like that, fortunately.
[08:08] mattjs76: I like physicists. They can be wacky sorts of people, and the more theoretical, they less obsessed they are with rigid fatalistic schemas.
[08:09] EsmayDean: Yep.
[08:09] EsmayDean: I quite agree.
[08:10] EsmayDean: But I've found open minded, interesting, inquisitive people in all the sciences. I just shake my head at the nasty and dogmatic ones.

(Matt Stinson, by the way, has a blog right here.)

Rumor has it that Google will announce its own Google PC on Friday, running its own non-Windows operating system. Of course, the source is given as the Los Angeles Times, well known to be one of the worst newspapers in America. But presumably their normal political biases wouldn't cause them to simply make something like this up.

If it's true I wish Google luck. I may well buy one of the machines in fact, depending on price and features.