• Google
    This Blog Web

October 2011

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          

RSS Feed

Bookmark and Share

Email Feed

  • Powered by FeedBlitz

« Politicizing Science | Main | Frustrating Discussion »

February 19, 2004


Feed You can follow this conversation by subscribing to the comment feed for this post.

michael vassar

Odd conclusion. I know from my my inquiries that molecular manufacturing is NOT accepted by the research communities of Drexel University, Penn State, or NIST, each of which have "nanotechnology" centers (one of which I have worked in). Nor is it accepted among those Florida State professors with whom I discussed the question. More strikingly, one physicist with whom I have spoken simply asserts that it is disreputable for a scientist to say ANYTHING about the feasibility of any technology that doesn't already exist or recieve major funding, citing the speculations of Gerrard O'Neil and Freeman Dyson as examples of what he called pseudoscience. I have also been at conferences where PhD'd scientists with "nanotechnology" grants discussed their perpetual motion designs (seriously). I'm glad Chris had a different experience, but I would like to know what he attributes the difference to. I would really like to see some reasonably detailed proposal to back up a claim like 5 "possible in 5 years with funding".

Brett Bellmore

Conference attendees are a self-selected group. It's scarcely suprising that people attending a conference on "Nanoscale Devices and System Integration" would be more open to the concepts of molecular nanotechnology than the general faculty.

michael vassar

As stated above, I have spoken to the faculty of "nanotechnology centers", not the general faculty. Ultimately the closed minds and lack of interest were part of why I gave up on academia.

Mr. Farlops

I think Brett has got the right of it. The conference is going to select for researchers with a particular mindset.

It sounds to me that Michael is dismissing MNT as empty and perhaps dangerous speculation.

Many scientists and engineers, even very respectable ones, have been guilty of vague speculations about future technology. Tsiokovsky speculated about spaceflight many decades before Sputnik. Turing speculated about artificial intelligence. Vannevar Bush, Sagan, Feynman, Tesla, etc. Granted that many of these speculations often come to naught, some are quite laughable and these scholars should know better, but they still do it anyway and sometimes, very rarely, they turn out to be right and they inspire others. How many engineers did Robert Goddard inspire with his vague speculations about trips to Mars and the Moon?

As one of the lay I'm certainly not qualified to judge, but MNT has this curious feel of "Oh yeah, that's obvious and doable. Why didn't anybody think this way before?" I imagine a lot of people working in rocketry or aviation in the early days felt like that--the science is sound, now it's really a matter of engineering.

Having said that, I'm still skeptical of the claimed proto-assembler in five years if major funding is granted. I think the attendees of the conference might be a little too enthusiastic. Or maybe Chris is projecting, I don't know.

michael vassar

I am not denying MNT's plausibility. It appears obviously workable to me as well, at least in its current form. The form presented in Engines of Creation was obviously workable but appeared to require either superintelligent AIs or many centuries to develop. More like DaVinci's mechanical speculation than like a serious plan. Now we have a plan however.
Speaking of DaVinci, it seems to me that someone who would buy his notebooks would also be interested in spending much more money persuing such a plan, especially if that person had much more money, the institutional influence to utilize it effectively, and at least enough interest in radical technological change to write a blurb for "age of spirituatl machines".

Greg Trocchia

It seems to me that the physicist that Michael talked to did not appreciate the fact that all speculations are not created equal. O'Neill's speculative book The High Frontier is less than 30 years old and his Physics Today paper upon which it was based is, IIRC, coming up on its 30 birthday. If you contrast this to Goddard's paper, A Method of Reaching Extreme Altitudes, which speculated about a moon landing, Goddard's paper (and was also ridiculed, famously, by the New York times at the time of its publication) reached its 30 birthday in the early ‘50s, when a trip to the moon was still Science Fiction (literally, Destination Moon was being made around that time). Were it not for WW II, which spurred the development of the V2, a trip to the moon would have been further away still. The fact that there is no Bernal Sphere in place at L-5 doesn't mean that spun gravity habitats and solar power satellites are "pseudoscience". It was and is clear that such things can be built, they simply are waiting on the cost to LEO dropping by an order of magnitude or more (as the shuttle was hyped to be able to do).

Contrast this to Kurzweil's Age of Spiritual Machines, which looks at the increasing computational horsepower available and compares it to a rough estimation of the computational capacity of the human brain and concludes that we are likely to see strong AI within the first half of this century. Now THAT, is speculation (though still not pseudoscience, I would contend. The issue of our ability to make use of all that computational horsepower is addressed nowhere near as convincingly and remains, IMO, the big question mark. Again, I am not saying that it won't happen, just that one can hardly take it as a given.

I see MNT as a middle case on this spectrum, but much nearer to O'Neill (and, by extension Goddard) than to Kurzweil. We face a tough "chicken or the egg" problem, how do we build devices capable of doing mechanochemistry without having mechanochemistry. A lot of work that has been done since the writing of Engines, including some done here, suggests that a bootstrap to mechanochemistry should be possible (though I think a talking about 5 years is a bit on the daring side).

I think that Drexler's tactical mistake was to write and publish Engines of Creation before writing and publishing Nanosystems. For one thing, Engines would have been better for his having dealt with the details of how to build MNT first. I doubt that you would have had the goo-like autonomous self replicating nanoassemblers depicted in Engines were this the case. I also think it less likely that Drexler would have got side-tracked into speculation concerning the potential of AI (in this particular area of the book, Drexler can rightfully be grouped with Kurzweil) which gave the misleading impression that strong AI is inextricably linked with MNT. Perhaps more important than this was the fact that when a skeptical technically-oriented reader was going through Engines, there would be references throughout to Nanosystems, showing that there was at least very solid theory behind the claims that he was making. I get the impression that by writing Engines first, Drexler made a bad first impression in certain parts of the scientific and technical community. For more on this, I recommend Ed Regis' book Nano (http://www.amazon.com/exec/obidos/tg/detail/-/0316738522/qid=1077369270/sr=1-3/ref=sr_1_3/1 04-5416049-1636738?v=glance&s=books

Brett Bellmore

I hardly think it was a mistake to publish Engines before Nanosystems. There are costs, too, of remaining silent until you've got an idea fully developed. And that would have been one long silence. I'd venture to guess that if he HAD waited that long, somebody would have beaten him to it.

In fact, I'd be very suprised if a lot of what you like about Nanosystems wasn't the result of constructive criticism of Engines.

Chris Phoenix, CRN

I should clarify: The people at the conference did not comment on my five-year estimate, because I didn't present that to them. And there were certainly some who were skeptical about the tech. I am certainly not claiming FIU as a hotbed of MNT fervor!

Re-reading my comments, I hope I don't give the impression that I'm citing FIU as MNT supporters in order to increase the plausibility of MNT. This was not my intention; MNT is plausible enough already. And I was mainly referring to people at the conference, some of whom are from FIU but many from elsewhere.

It's not that they support MNT--it's just that they don't have knee-jerk reactions against it, and they are generally willing to address it from an engineering angle, and most of them seemed to find what I said about it plausible. It was great to have five conversations in a row about MNT without hearing, "But what about thermal noise?" or "Sticky Fingers means you can't do the chemistry!" or any of the other tired old objections that have been debunked years ago.

So the takeaway message is: there's a significant population of researchers who seem ready, willing, and able to work on MNT, and who I think could develop it quite quickly if pointed in that direction. The real question is who will fund them first, and how soon.


The comments to this entry are closed.