Dean's World

Things are looking worse than ever for Hwang Woo-suk. It looks like the journal Science will withdraw his team's 2005 paper on human cloning, and may withdraw their 2004 paper as well, both of which were hailed as major advances.

This is an embarrassment not just to the field but to the journal and to the biotech profession in general. A growing concern in many areas of research, but particularly the biological sciences, has been that scientists have become so obsessed with headlines and money they're not doing the work they're supposed to do as diligently as they should, and the old tradition of others independently verifying a scientist's work by attempting to duplicate their results has become less popular than it once was.

Unlike some I don't think this means everything's wrong, but I've heard more than one real scientist, as well as educated layman, complain that the field has gotten too money and fame-obsessed, too quick off the mark with bold claims, with too much at stake in either success or failure for all involved, and real science itself winds up sufering.

This is a robot named Hubo. He's not a toy. He's been built by a South Korean group called the Korea Advanced Institute of Science and Technology.

Hubo weighs 55 kilograms and walks upright. He has voice recognition and voice synthesis, and hand articulation. I've had trouble finding video of him, but here's some video of Hubo using sign language. You can read more about Hubo here on Gizmodo.

Better-known than Hubo is probably Asimo from Honda:

He's been in the works the longest and has gotten the most exposure. He can not only walk, he can run, climb stairs, balance on one foot, talk, recognize individual voices and faces, and manipulate objects, all independently. Honda next year (that means 2006) plans to sell at least 100 copies for use in industry to do things like act as a receptionist and handle basic office drudgery.

For some amazing videos of Asimo in action, click here to see Honda's video clip collection. As you watch, contemplate: this isn't something that's wired to a human or has a human stuffed inside. This is an autonomous robot. He's not experimental, he's simply a prototype of a product that will soon be in production.

Then there's Sony's Qrio:

Qrio is closer to a toy since he's much smaller and doesn't seem to have the finger articulation of Hubo or Asimo. But just look at these amazing videos of Qrio in action. The thing can hop, balance on one leg, dance, throw a ball, catch a ball, follow motion, and even stand itself back up if knocked over. It's capable of doing all these things autonomously. More on Qrio on Sony's web site.

Now contemplate this: these things are going to be on sale soon. Yes, for ridiculously high prices. But given the law of accelerating returns, we should expect to see the prices plummet even while the capabilities of these things get more advanced.

By 2020 or so we'll have computers which are as complex as the human brain. That doesn't mean they'll work like human brains, but they'll be orders of magnitude more powerful than anything we have now, with far better advances in voice recognition, speech recognition, and so on.

The future is amazing to contemplate. This isn't geeky sci-fi stuff anymore. These are real-world phenomena that we are watching evolve in real life.

Update: Oh I totally forgot to mention this story about a laboratory robot that has demonstrated traits of self-awareness. No kidding. Check it out.

As a non-theist and a believer in Darwin (indeed, I'm a big fan of the entire field of evolutionary psychology), I have often been aghast at the fact that almost no one on my side seems to agree that most of the arguments over the supposed dangers of "Intelligent Design Theory" are not only foolish, but highly counterproductive and at times downright destructive.

Now along comes Michael Balter. He is a correspondent for Science Magazine, and serves as one of their chief sources on archaeology and human evolution. (For those who don't know, Science is one of the two most prestigious peer-reviewed journals in the world, the other being Nature, put out by our buddy Harvey Bialy's friends).

In writing on the "Intelligent Design" controversy, Balter asks a simple question:

"Could it be that the theory of evolution's judicially sanctioned monopoly in the classroom has backfired?"

If I said anymore I'd be gilding the lilly. Please click here to read Intelligent design and evolution, let's have a debate!

I invite you to leave your comments in response to Balter here, and I will bring them to his attention. However, while I shouldn't have to say this, I will:

If you leave a comment which makes it obvious that you have not clicked the above link and read Baltar's (quite short and quite eloquent) piece, I will not only delete your comment, but I will also press the secret button that only certain bloggers are given which blasts you with a 1.21 gigawatt thunderbolt right through your monitor screen. You have been warned!

I'm in the middle of reading Ray Kurzweil's The Singularity Is Near: When Humans Transcend Biology. I'm a skeptic by nature so when I heard about this book, at first it just sounded like overblown crankery, some kind of weird cult-y nonsense that overhyped the technological future. But as I heard more and more about it, the more interesting it got. So when Dave Price sent me the book for Christmas (thanks Dave!) I figured I had to read it.

So far what's impressing me about this book is not that it "blows me away" as the saying goes, but rather, just how nuts-and-bolts pragmatic and straightforward the man's reasoning is, and how well he backs up his arguments. He also explains very well what he means by "the singularity," noting that the word existed before black holes and simply meant something you couldn't see past. It's when a mathematical function essentially goes to infinity or otherwise behaves in a way that's difficult to understand.

What Kurzweil means by "the singularity" in terms of humanity and technology is that we will reach a point where technology has become so advanced, we today can no longer comprehend just how vast and far-reaching the changes will be--and he believes that this point will have arrived by no later than the middle of this century, if not sooner.

I'm still not finished with the book but at this point I'm convinced Kurzweil's no crank. I think what crystallized that for me was when I watched these videos of Asimo, the humanoid robot built by Honda. I suggest clicking the link and watching the videos, and contemplating the fact that Honda is actually making plans to mass produce these things and sell them as office assistants. Even if it's a few years before they're widely available, the fact is that it's only a few years. And just think about how much more advanced that little creature is going to get with continued research.

Kurzweil notes that in technology area after technology area, growth has been on an exponential path for over a century, and that trend is only accelerating in areas like computing, robotics, and biotechnology. If the growth stays on exponential curves that have typified the last 100 years, we will have computers which are as complex as the human brain within 15 years. They'll be wildly expensive supercomputers, but they'll be as complex as the human mind... and give it another ten years or so, and such computers will be ubiquitous and affordable by almost anybody. (Of course there's the question of how to write software for such a thing--but Kurzweil answers that too.)

Meanwhile, if we continue to learn at the same rate as we have about biotechnology, soon we'll be able to replace almost any organ, rework someone's biochemistry almost any way we want to, invent whole new organisms, or even directly enhance the human brain with implants.

The point at which all of these things are not just possible, but actually become commonplace is what Kurzweil calls "the singularity"--it's the point where all our technologies become so advanced that it profoundly questions all our notions of what it is to be human, and will change societ in ways we can only vaguely guess at.

And this is coming soon. Short of a major disaster it probably cannot be stopped.

One of the worst things you can accuse a scientist of doing is faking data he got published in a peer reviewed journal--especially one as prestigious as Science. Yet that is the accusation now facing cloning pioneer Hwang Woo-suk.

It's the sort of thing that, if proven, not only ruins a career probably forever, but may ruin the careers of several people around him forever.

If it's not true, this is a travesty of course.

Scientists are trying to create the first entirely artificial life form.

The eventual practical applications (if any) of this effort are currently unknowable, and will probably require a few more iterations of Moore's Law to fully understand and exploit as the information embedded in three-dimensional protein folding is as yet far too complicated to be modelled by even our most sophisticated processors and algorithms in a way that lets us have any hope of designing desirable attributes into a complex from-scratch organism, but if the random results of billions of years of undirected evolution are any indication, it seems to me this may be the beginning of the next great paradigm in technology, with human creativity once again bringing into reality things that we cannot imagine today but will not be able to live without tomorrow.

What do you think?

Here's an interesting story suggesting that some genetic testing of diseases may be unreliable.

It would not surprise me in the least. Indeed, it would not surprise me if a lot of things we think of as having a "genetic link" may not be all that strong a link to begin with. The human genome is only between 25,000 and 40,000 or so genes long, depending on what counting methods you believe and how much margin for error you'll accept. I suspect that long-term we're going to increasingly find that RNA and chromosomal structure and even things we don't understand at all right now are at least important than genes are. Future generations might look back on how quaint it was that we were so fascinated with genes.

Via Dan Riehl comes a genuinely interesting study of the subject of media bias, funded independently of any political or corporate donations, and using genuinely objective measurements.

Media bias is something I've always believed is real, if sometimes exaggerated. But among the story's more surprising finds are:

Special Report with Brit Hume is one of the few nightly newscasts that leans right at all, but, is also one of the closest to the center. It's no further to the right than ABC's "World News Tonight" and NBC's "Nightly News" are to the left.

NPR and PBS are far less left-wing than generally thought, and are actually admirably centrist.

If you exclude the editorial page, the Wall Street Journal is significantly to the left of the Washington Post or the New York Times.

Of course, by excluding all content that is opinion or non-news, this changes how they measure things. Yet it rings true: I often laugh out loud at how obviously biased to the left National Public Radio is, but I have generally found that when they're just reporting news, straight news, they tend to do it very well. Also, Fox News isn't foolng anyone, they're a right-wing news station all right, but, when they report straight news they are no more than a mildly patriotic, very very mildly right of center counterbalance to the likes of ABC News or NBC News.

This study's the first I've seen which uses objectively verifiable numbers too, taking rankings by Americans for Democratic Action and simply assuming that a score of 50% makes one centrist.

You can read details of the UCLA study here.

One of the more troubling aspects of modern medicine is the extent to which in recent years that people who write the medical guidelines, even the medical textbooks, for how to use certain drugs in clinical practice, also have a direct financial interest in the drugs they sell. One shouldn't have to think very hard before realizing that this creates a perverse incentive in the marketplace for medical care.

That doesn't mean doctors or researchers are evil people twisting their mustaches, of course. It's just something that would raise conflict of interest questions in any business--and medicine is a business, never kid yourself about that. Indeed, one of the more destructive things we've done as a nation is to act like doctors are somehow above the normal human foibles; the truth is that the patient-doctor relationship is a client-service relationship. This is even more true with faceless entities like the drug companies. People who forget that do so to their own detriment.

This statement will be called "demonizing the drug companies" by some. No, it isn't. Noting when someone has a conflict of interest isn't demonization. It's normal ethical business practice. It's also normal ethical scientific practice.

We wouldn't accept the notion that for a fee your prosecuting attorney could switch sides and become your defender, or vice versa. Yet we often allow people with direct financial stake in certain drugs or tests to also be in charge of most, or all, of the research and clinical practices associated with them. That fact is rarely disclosed to patients. Yes, the FDA is there to prevent the most gross abuses by forcing certain minimum standards, but that's as subject to politics as anything else.

I found this British Medical Journal on disease-mongering quite eloquent on the matter.

A new study finds that the earliest Americans may not have been the same racial group as American Indians. Which is a political hotbutton issue for some.

Daily Pundit notes similar articles raising the issue.

Quoted:

Hi Dean Esmay,

I'm writing to share a recent paper of mine that I thought you might be interested in...

British Medical Journal -- Dec 10, 2005
BMJ 2005;331:1412 (10 December)

"Are US flu death figures more PR than science?"
http://bmj.com/cgi/content/full/331/7529/1412

If you post a "rapid response" to my article on the BMJ website, please do send me a copy.

And certainly any comments you might have about the paper would be most appreciated!

Regards,
Peter

I am, as the old saying goes, so busy and so stressed at the moment I don't know whether to poop or go blind, but I thought it was interesting enough to pass on to Dean's World readers, who I'm sure will leave some interesting comments.