For those that don’t know, the Foresight Institute is an organization devoted to promoting the upside and avoiding the dangers of transformative future technologies. Their primary focus is on nanotechnology and it’s ultimate expression: molecular manufacturing as expressed in Feynman’s “There’s Plenty of Room at the Bottom” talk. And of course Drexler’s Engines of Creation developed these ideas further. So this is pretty technical stuff and the Foresight 2013 conference was the most academically focused futurist conference I have attended. Several attendees even remarked that it was more academic than previous years. I attended in 2010 for the great Moldbug/Hanson debate and again in 2011 at Google when they had a more entrepreneurial focus. I guess those were less academic. But my brain melted slightly in the blast furnace of atomic scale physics each time.
I like to challenge myself with these things, trying to absorb some fractional understanding of the work presented by these top scientists in highly specialized fields. People looked at me askance previous years when I told them I was just a layman interested in the field. This year I lamely suggested that I was a blogger of some sort and found that this was not more ingratiating given the media ban. Fortunately for me, I am shameless and slightly pushy in conversation so I manage to get my ears filled up with some amazing ideas even if I do barely comprehend them. Being a generalist, I am biased toward the idea that everyone is a laymen in fields other than their own. I hope that I can help pollinate ideas across fields with my writing some day.
Given my interest in computers, one of my favorite Foresight presentations so far this year was a talk by Purdue professor, Gerhard Klimeck about single atom transistors. Luckily, I found a similar presentation already posted on Nanohub.org, so I will talk about that. One key point worth noting is that cpu performance is really constrained by power consumption. We get more transistors but clock speed and performance per clock cyclehas been pretty much flat since 2005. Which is why parallel computing is so important. But in spite of the nifty .NET tools mentioned in the link above, parallel code is still harder to write and largely under-utilized. This is something that Paul Graham has carped about with his ambitious startup idea: Bring Back Moore’s Law. And of course it smacks of Theil’s stagnation of innovation schtick. But I’m sure Ray isn’t worried. He knows aparadigm shift will save the day.
So wait, oh yeah, power consumption occurs when circuits are switched and via leakage. In fact, as much power is supposedly lost via leakage as from switching. So your CPU is constantly leaking juice. Disgusting. So these single atom transistors come riding to the rescue since they have less leakage. But Klimeck’s main contribution to this effort is theNemo5 software which enables researchers to model “atomic-resolution calculation of nanostructure properties.” Modeling is a key enabler of all design I guess, and this Nemo software seems to have a broad range of uses from academia to industry. Good on him and his group. Where would our Singularity be without them?
I really wish I could talk more about my amazing conversations tonight, but I will just link to this one paper which is already public but whose significance has not yet been widely appreciated: Neutral Atoms Behave Much Like Classical Spherical Capacitors. If you are a super-genius, I assume it will be obvious to you why this is important. Listening to this idea connected to a bunch of other ideas gave me some glimmer of insight into the matter, but it will all be made more clear by forthcoming publications. Stay tuned to your physics news feeds, friends.
One of my favorite speakers today was Northwestern researcher, Bartosz Grzybowski. He was a pragmatic and plain spoken presenter. I can’t discuss his presentation, so I will just link to his team page for now. I chatted with him briefly and we agreed that the current quality of science journalism leaves something to be desired. It seems to me that there is constant pressure on both news services and academics to promote new findings. I would further suggest that many of these findings are being oversold by academics and under scrutinized by journalists. Grzybowski referred me to a blogger that he likes namedPhilip Ball and Ball’s article on the hype around graphene offers a good example of critical science journalism. So we just need to make sure the rest of our science writers getRoyal Society Fellowships before we pay attention to them.
I was also impressed by the “Self Organizing and Adaptive Systems” session chair Lee Cronin from the University of Glasgow. Cronin was an outspoken critic of the poorly defined buzz words and jargon being generated in his field. He also gave a presentation of his own which will not be blogged about by me here and now. But his lab does appear to be employing some advanced automation as evidenced by their Network Flow Systems and their interest in commodity 3d printers. I am always excited whenever I see commodity stuff like the hackerspace prereq 3D printers opening up new applications due to reduced cost. The SingInst folks also have a cool page on automated science, and I must say that it’s reassuring to see that scientists face the same threat of obsolescence asretail clerks and -err- Motorhead cover band members. Ha, just kidding, we all know that the true skill of any scientist lies in their mastery of the black art of grant writing and we don’t have good enough NLP to replace that yet.
George Church closed out today’s session with a bunch of mind blowing collection of DNA breakthroughs. Douglas and Bachelet’s DNA nanorobot can carry a therapeutic payload which is only unlocked by a target such as a cancer cell. (Already published so free for me to discuss.) Nanopore sequencing has apparently been dubbed the winning DNA sequence technology and has put Halcyon off their feed. But I am sure those Halcyon guys have more tricks up their sleeves. DNA is also good for data storage apparently. This CRISPR gene editing technique looks promising. Wait, gene editing? When did this happen? Oh sure, zinc fingers have been around for a while. I have to admit this stuff has the SciFi writer in me coming up with conspiracy theories.
This article is a continuation of my commentary on the Foresight 2013 conference. As I mentioned in my Day 1 and Day 2 posts, the Foresight folks have a strict media policy in place. So while I can’t really blog about the content of the presentations, I will discuss the work these speakers have previously made public.
I am not a morning person (I’m a b-chronotype for those in the know.) So I was prevented from seeing the morning speakers by the tyranny of the early risers. But I did enjoy seeingArthur Olson of the Scripps Research Institute, which is a private, non-profit, biomedical research organization. Olson’s work focuses on modeling processes at the molecular level. He was a pioneer of molecular graphics and his lab is appropriately called the Molecular Graphics Laboratory. You should definitely take a look at some of the “Tangible Interfaces for Molecular Biology” demos. Super cool.
These tools allow researchers to turn molecules over in their hands and examine the structure or even attach side-chains and basically play with the models. Olson joins manyother researchers in his realization that play is incredibly important to the learning process. He pointed out that Crick and Watson used a physical model to refine Franklin’sdiscoveries from her DNA x-ray diffractions. Physical models provide a rich cognitive substrate and aid in social cognition. Olson extends his work with physical models into education with his Science Within Reach project. The amazing Inner LIfe of the Cell video comes to mind with regard to Olson’s work. Apparently the Harvard students who watched this animation scored much higher on a quiz about these cell functions than students that didn’t get to see it . Sucks to be in the control group sometimes.
Anyway, Olson’s group does a lot of other cool stuff like ligand-protein docking modeling with Autodock. I got confused about this whole Autodock force field idea. I guess for chemists, its more about potential energy approximation than about, you know, protecting astronauts or something. They also work on an embedded Python Molecular Viewer (ePMV) that “runs molecular modeling software directly inside of professional 3D animation applications” like Maya or Blender. This like hijacking the tons of money poured into Hollywood-style animation tools and putting it to work for science. Awesome. Somecranky types like to point out that models misrepresent and oversimplify the complexity of reality. But screw those guys if they can’t grasp how powerful (and beautiful) these pretty close representations are.
James Ellenbogen from the MITRE corporation also spoke at Foresight 2013. Ellenbogen heads the MITRE Nanosystems group and runs the student program there. One of his more prominent accomplishments recently was his involvement in the creation of the first programmable nanoprocessor. Now this is a cool thing to have a computer no larger than a human cell. Yeah, slap a fractal antenna on that nanoprocessor and go all Fantastic Voyage. But Ellenbogen’s presentation had such an amazing ending (which I can’t discuss) that it sort of takes the wind out of my sails. Just seriously watch those physics news feeds, I’m telling you.
Ellenbogen’s student program sounds interesting. A bunch of wunderkind go off to a little known private research corporation to work on secret projects of national importance. Where’s Dr. Xavier? Also, what’s with this MITRE place anyway? I guess it’s some sort of sister entity to the RAND corporation with a focus on complex network systems. They build a lot of stuff for the FAA. When I was trying to find info on MITRE, I quickly came across a bunch of posts by the tin-foil hat crowd who figure MITRE must have been in on 9/11 since they have root access to the air traffic control systems. Oh yes, it’s all falling into place now.
Stay tuned for the final Foresight installment: Alan Aspuru-Guzik on what quantum computing is really good for, Gerhard Klimeck shares knowledge worldwide with Nanohub.org, and Ron Dror and William Goddard give us yet more reasons to respect their models.
I would love to say that anyone who thinks they understand quantum mechanics doesn’t understand quantum mechanics, but I really just don’t understand it. When Harvard’s Alan Aspuru-Guzik gave his Foresight 2013 talk “Simulating Quantum Mechanics with Quantum Devices,” I listened with more enthusiasm than comprehension. So bear with me. Aspuru-Guzik likes to use quantum simulation to go after electronic structure calculations which are some of the most computationally intensive problems in science. ”The calculation time for the energy of atoms and molecules scales exponentiallywith system size on a classical computer but polynomially using quantum algorithms.” Aspuru-Guzik points out that theory is ahead of experimentation in this field, but he has found and built some toys to play with.
So the idea here is to leverage quantum devices to simulate quantum mechanics. I guess the NIST has some device with hundreds of qubits. But the systems Aspuru-Guzik gets to play with are more modest. He ran a simplified protein folding problem on an 81 qubit D-Wave system and got 13 correct results out of 10000 runs. ”The fact that it worked at all was significant.” The investors must be thrilled. I have heard that aside from factoring numbers, there aren’t many uses for this quantum computing. But if you can factor numbers, you basically break all encryption. Of course when I say “you” I mean the NSA. But Aspuru-Guzik’s stuff is more benign. He will be folding proteins and figuring outphotosynthesis and stuff. So he’s cool.
Next, Gerhard Klimek gave a talk about Nanohub.org. Here’s what they say about themselves:
WHAT IS NANOHUB.ORG?
nanoHUB.org is the place for computational nanotechnology research, education, and collaboration. nanoHUB hosts a rapidly growing collection of Simulation Programs for nanoscale phenomena that run in the cloud and are accessed through your web browser. In addition there are Online Presentations, Courses, Learning Modules, Podcasts, Animations, Teaching Materials, and more to help you learn about the simulation programs and about nanotechnology. nanoHUB supports collaboration via Workspaces and User groups.
So there are clearly educational resources for students, but I understand that researchers and industry folks get into the simulation stuff. Boasting 900 papers with an h-index of 41, Nanohub is a serious scientific resource. So why head head on over and simulate a carbon nanotube or something?
Carrying on in the simulation vein, Ron Dror of D.E. Shaw Research talked about their custom supercomputer, Anton. Anton is a massively parallel ASIC based pocket calculator that can figure out how drugs bind to receptors. Dror has published work on G-protein-coupled receptor modulators in particular, which represent one third of all drugs. Who knew? Pretty cool stuff. And this David E. Shaw fellow is an “intriguing and mysterious” character. He saunters from his Stanford PhD over to Columbia, toys with parallel supercomputing, yawns, strolls down to Wall Street, dabbles with high frequency trading, stretches, casually sets aside the resulting $27 billion hedge fund and sets up a computational biochemistry research group to model molecular dynamics simulations of proteins. What a slacker.
Topping off the conference was the venerable CalTech theorist, William A. Goddard, III. Your guess is as good as mine as to what he said… and I was in the audience. There was something about a ReaxFF force field which lets you model chemical reactions. He also said he was happy to see theory starting to be able to predict something useful, which I am sure is a huge understatement. But there was just too much math for me to really get a grasp on his talk.
I was incredibly awed by these sober scientists toiling away at the edge of human knowledge, delving into the the very underpinnings of chemistry and biology.What new wonders will be within our grasp as we come to understand and manipulate complex molecular interactions at the atomic level? Dare I hope for my beloved utility fog someday? If so, we will have them to thank. And uh, possibly pay royalties to, depending on how the IP plays out.
[This article originally appeared in Scott's Oakland Futurist blog: http://oaklandfuturist.com/foresight-2013-day-1/ , http://oaklandfuturist.com/foresight-2013-day-2/, http://oaklandfuturist.com/foresight-2013-day-3/, and http://oaklandfuturist.com/foresight-2013-day-3-part-2/]