Michael Graham Richard

Q: If we assume that both hands above came from an honest deal (truly random), which one are you more likely to get?

Read the rest of this entry »

In the past decades, we’ve all heard about the progress made in organ transplant science and in the therapeutic cloning field, but advances in artificial bones have rarely made waves.

Bones are very light but nonetheless able to withstand extremely heavy loads. The inside of a bone is like a sponge. It is particularly firm and compact in certain places, and very porous in others. [...]

Researchers at the Fraunhofer Institute for Manufacturing Engineering and Applied Materials Research developed a simulation program that calculates the internal structure and density distribution of the bone material. [...]

Engineers can produce complex components with the aid of rapid prototyping technology. This involves coating a surface with wafer-thin layers of special metal powder [made of biomaterials such as titanium and steel alloys]. A laser beam heats – or sinters – the powdered metal in the exact places that need to be firm. [...]

“The end product is an open-pored element,” explains [Andreas] Burblies. “Each point possesses exactly the right density and thus also a certain stability.” The method allows the engineers to produce particularly lightweight components – customized for each application – that are also extremely robust.

Of course, the ultimate goal is to replace bones as rarely as possible, and in the long-run that can probably be accomplished with complete rejuvenation therapies (how many people in their 30s need to have bones replaced?), but until we get there and for special cases, the best alternative is to swap out those worn out knees for the best reproductions possible.

And while we’re at it, why not replace bones with superior artificial bones. Lighter, stronger, etc. No reason to limit ourselves to what evolution gave us.

Sources: Fraunhofer-Gesellschaft, Science Daily.

A few days ago, I was laying in bed, trying to fall asleep without much success. After reading for a while, I turned off the light and ended up on my stomach, with my head resting on my folded arms. I was getting sleepy, so I decided to get in a more comfortable position before I fell asleep. But right before I could physically execute my decision, I thought about how I was going to do it.

Not as easy as it sounds. I knew for certain that I could get out of that position - I’ve probably done it thousands of times - but I just didn’t know precisely how I would do that in advance. How would I move? Would I shift my weight this way, or that way? Support myself on my right elbow? Try to spin and then use my back muscles to sit up? I had a vague idea of how it went, but the details truly stumped me, and I just laid there trying to figure it out.

Why is it so hard to predict in high-resolution totally mundane movements that we’ve done a thousand times?

Is it because we can do most of them without paying attention, and so we’re simply not used to thinking about it. Maybe with practice I could improve?

Is it because there’s an evolutionary advantage to not cluttering our conscious mind with it, so we’re not equipped on the hardware level to precisely and reliable forecast complex - if banal - physical movements?

Or maybe it’s because even the motor centers of our brain don’t know the details in advance, and rely on many feedback loops to adjust things in real-time once the sequence has started? After all, even when doing familiar things, you never move in exactly the same way, so there’s got to be some room for improvisation and adjustments.

Read the rest of this entry »

What is metabolomics?

Genes are similar to the plans for a house; they show what it looks like, but not what people are getting up to inside. One way of getting a snapshot of their lives would be to rummage through their rubbish, and that is pretty much what metabolomics does. [...]

Metabolomics studies metabolites, the by-products of the hundreds of thousands of chemical reactions that continuously go on in every cell of the human body. Because blood and urine are packed with these compounds, it should be possible to detect and analyse them. If, say, a tumour was growing somewhere then, long before any existing methods can detect it, the combination of metabolites from the dividing cancer cells will produce a new pattern, different from that seen in healthy tissue. Such metabolic changes could be picked up by computer programs, adapted from those credit-card companies use to detect crime by spotting sudden and unusual spending patterns amid millions of ordinary transactions.

This could be used for traditional medicine, both to prevent pathologies and to detect those that are already present so they can be treated. But another use would be as part of an early-detection system to defend against pandemics and biological attacks. As mentioned previously, network-theory can help us better use vaccines. But once you have a cure or antidote, you also need to identify people that are already infected but haven’t died yet, and the earlier you can do that after the infection, the more chances they have to live.

Once the techniques of metabolomics are sufficiently advanced and inexpensive to use, they might provide better data than simply relying on reported symptoms (might be too late by then), and might scale better than traditional detection methods (not sure yet - something else might make more economic sense). It’s a bit too early to tell, but it’s a very promising field.

For more information, see Douglas Kell’s site or Wikipedia: Metabolomics.

Source: The Economist. See also Lifeboat’s BioShield program.

This was cross-posted on the Lifeboat Foundation blog.

I just got back from a concert at the National Arts Center in Ottawa. It was the NAC orchestra led by Pinchas Zukerman. More details here.

When we got to our seats, I noticed that we were sitting next to an Asian-looking couple. What immediately popped to mind was “Oh yeah, not surprising, classical music is relatively popular in Asian culture right now” because of some articles that I read about how Asia is training more classical musicians than anywhere else, and that “Western” classical music might have a brighter future there than in this hemisphere.

But that was just confirmation bias.

If I had really wanted to look at this more rationally, a good start would have been to find out what portion of Ottawa’s population is of Asian origins, and then to look at what portion of the concert’s spectators was of Asian origins. I don’t know these numbers, but it might very well be that there were less Asians in the room as a percentage than their ratio of the local population. This could mean a lot of things, like that people of Asian origins who don’t live in Asia anymore aren’t more likely to be classical music fans than the general population, or that age is a bigger factor (lots of white hair in the room) than cultural background. Etc.

Bottom line, just remember to check for confirmation bias often. You get better at catching yourself with time, and someday it might help you avoid a bigger mistake than having an inaccurate belief about Asian culture and classical music.

See also: Rationality.

If a pandemic strikes and hundreds of millions are at risk, we won’t have enough vaccines for everybody, at least not within the time window where vaccines would help. But a new strategy could help use the vaccines we have more effectively:

Researchers are now proposing a new strategy for targeting shots that could, at least in theory, stop a pandemic from spreading along the network of social interactions.Vaccinating selected people is essentially equivalent to cutting out nodes of the social network. As far as the pandemic is concerned, it’s as if those people no longer exist. The team’s idea is to single out people so that immunizing them breaks up the network into smaller parts of roughly equal sizes. Computer simulations show that this strategy could block a pandemic using 5 to 50 percent fewer doses than existing strategies, the researchers write in an upcoming Physical Review Letters.

So you break up the general social network into sub-networks, and then you target the most important nodes of these sub-networks and so on until you run out of vaccines. The challenge will be to get good information about social networks, something not quite as easy as mapping computer networks, but there is progress on that front.

In one of the most dramatic illustrations of their technique, the researchers simulated the spread of a pandemic using data from a Swedish study of social connections, in which more than 310,000 people are represented and connected based on whether they live in the same household or they work in the same place. With the new method, the epidemic spread to about 4 percent of the population, compared to nearly 40 percent for more standard strategies, the team reports.

Source: ScienceNews. See also Lifeboat’s BioShield program.

This was cross-posted on the Lifeboat Foundation blog.

In March, Technology Review publish a piece about some enzymes that the lab of Dr. David Baker (the head of Rosetta@home, one of my favorite scientific distributed computing projects) designed from scratch.

In a major step forward for computational protein design, scientists have built from scratch a handful of enzymes that successfully catalyze a specific chemical reaction. These proteins have no naturally occurring counterparts, and the reaction–which breaks down a man-made chemical–has no natural catalyst.

As you can see from the comparisons between catalyzed and un-catalyzed reactions in the graph above, when it comes to enzymes, it’s all about speed.

Of the 72 proteins selected, 32 successfully helped along the reaction. The most efficient proteins sped up the reaction to 10,000 times the rate without an enzyme.

While that’s an impressive feat compared with earlier enzyme design attempts, the synthesized enzymes pale in comparison to naturally occurring ones. “It’s not very good at all,” says Baker. “Naturally occurring enzymes can increase the rate of reactions by much, much greater amounts”–as much as a quadrillion-fold.

Read the rest of this entry »

Here’s another fascinating update about my Chili plant. I first wrote about it last Autumn, then gave you an update about my hand-pollinated winter Chili crop.

I thought that 7 peppers at once was a pretty good harvest. Boy, was I wrong. Above is a photo I just took of the very same plant. It now lives in my parents’ garden and it looks like it’s about to collapse from the sheer weight of all those chili peppers. Incredible.

Read the rest of this entry »

I was recently reading my copy of The Economist and there was this reference to the Materials Genome Project:

The 30,000-compound question
At the moment the process of finding better electrode materials is haphazard, but Dr [Gerbrand Ceder, a battery scientist at MIT,] proposes to make it systematic. Over the centuries, chemists have discovered about 30,000 inorganic chemical compounds (those that are not based around carbon skeletons), almost any of which might theoretically be suitable material for an electrode. Examining the relevant properties of all of them in the laboratory is out of the question, but Dr Ceder thinks he has found a short cut. He is involved in something called the materials genome project, which takes the known properties of inorganic compounds and turns them into extremely sophisticated computer models. These models are able to calculate the quantum-mechanical properties of the chemicals they are mimicking—and they seem to get it right. When Dr Ceder has checked the predictions for hitherto untested materials by conducting real experiments, he has found that the results coincide.

It’s a brilliant idea! I would love to see them try a distributed computing approach to speed things up and keep costs down. Since potential benefits to humanity are so great, they wouldn’t have problem finding volunteers to donate CPU cycles.

Update: I have received a email from Dr Gerbrand Ceder, the man mentioned in the Economist piece, and apparently the website that I originally linked is for a different project with the same name. I’ve removed it to avoid confusion.

On Friday, June 27th, leading scientists and thinkers in stem cell research and regenerative medicine will gather in Los Angeles at UCLA for Aging 2008 to explain how their work can combat human aging, and the sociological implications of developing rejuvenation therapies.

Aging 2008 is free, with advance registration required at http://www.mfoundation.org/Aging2008/

There’s a press release here, and you can find more information on the official Aging 2008 website.

If you are not familiar with longevity science, Aubrey de Grey’s TED talk is a good starting point. If that intrigues you, his book Ending Aging is the next logical step.

Update Great news! I email Jeff Hall, the coordinator of Aging 2008, to ask if videos of the event would be made available online. He answered that they will “be filming the entire event and putting it up on youtube.”

I wouldn’t normally do this, but my Fold It: The Protein Folding Game post generated quite a bit of attention, and it is obvious that many people are curious about the protein-folding game. So I figured that a central repository of media coverage might be helpful - more information might push a few people over the edge and convince them to try the game. Here’s what I’ve found so far:

• University of Washington Official Announcement (which was reprinted on Science Daily and Physorg, among others): Computer game’s high score could earn the Nobel Prize in medicine
• Howard Hughes Medical Institute Announcement: Researchers Launch Online Protein Folding Game
• The Economist: Return to the fold
• Technology Review: Biologists Enlist Online Gamers
• Slashdot (misleading title): Folding@Home 2.0 - An Online Protein Folding Game
• Penny Arcade (last paragraph): Je Parle Un Peu
• Nature Magazine Blog: Addictive protein folding game
• Seattle Times Blog: Free video game from UW: Wiggling and shaking for science
• Kotaku: Foldit Makes Protein Folding A Game
• Seattle PI: Scientists tap gaming’s power

If you see Fold It mentioned in the mainstream media or on a popular website, let me know in the comments and I’ll add it to the list.

Fold It is a game developed by the Rosetta@Home team (learn more about distributed computing) under the direction of Dr. David Baker at the University of Washington.

It features a new approach to protein prediction. Instead of using a more or less brute-force approach, with a CPU trying lots and lots of possibilities and calculating which ones give the best results, the game uses the human brain’s pattern recognition abilities (with help from a few automated tools) to try to find the lowest-energy folded state of a protein.

It has the potential to be on the cutting edge of a new generation of scientific games that are fun to play, teach you things, and can actually help researchers.

Words are inadequate to describe it, so please watch the two videos below to get an idea.

Read the rest of this entry »

A recent article in Technology Review by Nick Bostrom generated a lot of discussion about the Fermi paradox, which states:

The size and age of the universe suggest that many technologically advanced extraterrestrial civilizations ought to exist. However, this hypothesis seems inconsistent with the lack of observational evidence to support it.

I’ll add my 2 cents to this discussion by saying that there’s a possibility that any civilization that becomes advanced enough discovers that physical reality can’t hold a candle to virtual reality and makes the transition (alien transubstantiation, to coin a phrase). This could explain why they haven’t colonized the galaxy, or why we aren’t bathed in their radio communications.

Virtual worlds can be, in theory, both much more pleasant to inhabit, with unlimited freedom and none of the downsides of an existence based on crude physical processes, and also much more energy-efficient. Even without cold computing, it would take a lot less energy for an advanced civilization to do all that it wants to do within a simulation than by moving atoms around.

As I mentioned before, they could also think much faster, subjectively pushing back the heat death of the universe (while at the same time making communication with ’slow’ beings almost impossible).

I haven’t read all the serious papers on SETI and the Fermi paradox yet, but I’m pretty sure this is not an original theory. It’s just something that I haven’t seen mentioned yet and that I think deserves thinking about.

Update: Just to make things clearer, the kind of virtual reality I’m envisioning here is not one where you connect a biological body to a machine that sends it sensory information (like in the Matrix, for example). What I’m thinking of could probably be called ‘mind uploading’. There is no physical body, because one is not required. Everything would be inside the virtual world, kind of like how an artificial intelligence would not require a physical presence other than its computing substrate.

Welcome Ars Technica visitors: If you liked this, you can subscribe to my feed here.

For the past few days I’ve been reading (among other things, of course…) Meditations by Marcus Aurelius, a roman emperor who lived from 121 to 180. He is known as one of the most important stoic philosophers.

One thing that has been on my mind while reading this is the fact that many people are very impressed by anything labelled “ancient wisdom” and have a bias towards giving it more weight than more recent thought. Part of that inclination is rational: If something has endured that long, there’s a good chance that it is because of its quality. But another part of it is not rational. It is based on the false parallel between the fact that older humans are generally considered wiser and the fact that the text is old.

From our point of view, the text is old. But from the point of view of human knowledge, old texts are ‘younger’ than modern texts.

So while I appreciate many of Marcus Aurelius’ stoic principles (look for truth, mind your own business, don’t waste your time on frivolous things, clearly define what matters to you so you can better stick to it, be open to have your mind changed by evidence, eliminate the unnecessary, etc), I simply chuckle when I read about his conception of the universe, the gods, reality, destiny, dualism (soul separate from body), death, etc. This is the best information that was available at the time, but compared to what we know now, it’s clearly archaic and if the roman emperor had been born today, he probably wouldn’t believe what he believed then (not to mention his positions on slaves, women, homosexuals, etc).

Yet some people will automatically give more weight to these ideas than to ideas that come from more contemporary sources because they come from “ancient wisdom”. If you suffer from that bias, you should recognize it, look back on how it might have influenced you in the past, and keep it in mind for the future. Judge ideas on their own merit, not on their capacity to endure the passage of time. With some things, it doesn’t matter too much (f.ex. morality). With others, it changes everything (scientific fields such as cosmology, biology, physics, etc).

If more people realized this, fewer Bronze Age myths would be taken seriously.

Here are 7 questions that I would like to ask to the following people: Michael Anissimov, Jamais Cascio, Tom McCabe, George Dvorsky, Steven Smithee, Randall Parker, and ‘Reason’ of Fight Aging.

Guys (no girls in my blogroll, sadly), you don’t have to reply if you don’t feel like it, but if you want to, just post the answers on your blog (I’ll link back to the entry) or in the comments here. Anybody else who wants to participate by answering one or many of the questions is welcome to do so in the comments. Thanks!

1. What would you nominate as the best idea that anybody has ever had? Why?

7. Looking ahead, are you an optimist or a pessimist? Why?

Update: ‘Reason’ from Fight Aging has answered my questions here. Michael Anissimov has given his answers in the comments below (keep an eye out for his upcoming book!), as well as Jamais Cascio. Someone going by the name of ‘Infidel753′ gave his answers over on his blog. Also in the comments are answers by Jeremy Sheperd, Dustin Parsons, and Zach. George Dvorsky also replied on his blog. Many thanks!

My own answers can be found below.

Read the rest of this entry »

This story doesn’t seem to have been picked up by the media yet (and maybe it won’t, but these kinds of things usually are because we like to know in which activities we can still beat machines):

PARIS, April 9 /PRNewswire-FirstCall/ — During the Go Tournament in Paris, staged between 22 and 24 March 2008 by the French Go Federation (FFG), the MoGo artificial intelligence (IA) engine developed by INRIA - the French National Institute for Research in Computer Science and Control - running on a Bull NovaScale supercomputer, won a 9×9 game of Go against professional 5th DAN Catalin Taranu. This was the first ever officially sanctioned ‘non blitz’ victory of a ‘machine’ over a Go Master.

That’s impressive on its own, but lets not jump the gun and claim that humans have been defeated at Go in the way that they have been defeated at Chess (Kasparov v. Deep Blue, 1997). The reason for that is that a board game usually has 361 squares (19 x 19), and this one only had 81 (9 x 9).

That makes quite a big difference in the size of the ‘tree’ of possible moves that has to be searched by the computer, which means that the brute force approach is less effective and so the software has to be smarter in how it approaches strategy and tactics to perform well.

Not surprisingly the Go master beat the computer in a game on a 19 x 19 board, and that with a nine-stone handicap. But even after that, he was impressed:

[...] the Go Master nevertheless rated the IA system as ‘approaching Dan standard’ in a performance that promises some formidable battles to come between man and machine.

Formidable battles indeed! The final one will be fought over Artificial General Intelligence (AGI).

Source: Latest Advance in Artificial Intelligence: Computer Wins a Game Against a Go Master

I was reading and article in The Economist about lasers that can pulse extremely rapidly. We’re talking really fast, in the femtoseconds range (one billionth of one millionth of a second).

This got me thinking about the nature of time: Is there a theoretical limit to how fast something can happen? I’m not aware of any, but physics probably gives an answer one way or the other.

Still, even if there’s a limit somehow, there’s still quite a gigantic range. From femtoseconds to how long it takes for universes to die.

What if what we consider to be “real time” - how fast we move, talk, think - happens to be a glacial pace compared to other lifeforms? I’m not sure if biological intelligent life could have a subjective impression of time on such a scale because of limits to the speed of chemical reactions and the minimum complexity required for intelligence, but if an advanced civilization had made the transition to a non-biological substrate (such as super-computers), it would be conceivable that for them seconds could subjectively be the equivalent of millennia (or more) to us.

That would make communication unlikely. It would be a bit like trying to have a conversation with a rock. Even if you knew it was intelligent, you’d probably be bored out of your mind and either you would ignore it, or wait for it to speed up. And even if that’s too anthropocentric a way to look a the situation, there’s still the problem of saying something coherent mentioned below.

There’s always the possibility that such a fast intelligence would remembers how slow it once was, in its original bio-chemical form, and plan for future contact with lesser intelligences. Keep listening on the ’slow lane’, in other words. But even if it did that, could it really communicate with us coherently if between each syllable it had the time to evolve and change a lot (more than Homo Sapiens has had time to evolve so far)? Even if it creates the message in its ‘real-time’ and then slows it down to send it, will the entity that created the message have much in common with the subjectively much older entity that exists by the time the message has been completely sent?

A few months ago I discovered Futures in Biotech, a podcast that “explores the world of genetics, cloning, protein folding, genome mapping, and more” by Marc Pelletier. The episode that first caught my attention was an interview with Dr. Vijay S. Pande of Folding@home. In fact, I might have found Futures in Biotech via a link on the Folding@home blog.

Anyway, I had a look at their archives and saw that they didn’t have anything about Aubrey de Grey or the Methuselah Foundation. I figured it would be a perfect fit and that their listeners would be interested by the SENS platform.

So I found Marc’s blog and I either emailed him or left a comment (can’t remember) suggesting that he interview Aubrey de Grey, and gave some links to check out.

Today, I returned to Futures in Biotech and was very happy to see that their latest episode is an interview with Aubrey de Grey! I haven’t listened to it yet, and I can’t be sure that it was my comment/email that was the seed from which this grew, but whatever the cause was, I’m quite happy that a few more people will be exposed to SENS.

Maybe some of the listeners of that show are biology students or research scientists and this will start a chain of events that will lead them to help healthy life-extension research directly or indirectly, or maybe some of the listeners will donate to the Methuselah Foundation. This can only help.

You can listen to the interview here.

Update: I’ve now listened to the interview and it’s quite good. Unfortunately, the sound quality for Aubrey’s side of the conversation isn’t very good and I missed some of what he said.

I encourage you to listen to the interview, but know that it is not the best introduction to the SENS platform. A better start would by this TED talk, and the most complete and detailed overview is Aubrey’s book.