• 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

« Sunday Afternoon | Main | Reality Check »

May 01, 2005


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


I see space colonies as a fail-safe for many earth-centric human extinction scenarios, including some of the MNT enabled ones; probably not far enough away to be safe from an "evil" AI though. The only estimate I've seen for payload requirements of a moon-colony is 1000 tonnes, about half of an SE's annual capacity. If there are any potentially dangerous MM experiments to be performed in the time after MNT is reached, it would make sense for the lab to be in an isolated locale, such as space.


The 21st Century is NOT going to be a wild ride, but rather a sluggish crawl. I don't know where you get the 10-20 year estimate for molecular manufacturing from, but one thing's for sure, I don't hear about much progress from reading the news on this site, KurzweilAI, Small Times, NanoApex, NanotechnologyNow, or any of the other sites. Progress is happening at an incremental rate, NOT exponential as you claim. The whole nanotech "industry" is EXACTLY where it was in 1990. I don't see anything on the horizon regarding a molecular assembler, so I doubt we'll see one in action until 2060-2075 at the earliest.

Michael Laine

ambition, not to mention audacity...

thats maybe the nicest thing some people have said about it! some people have called it a wasted effort, some call it a scam. for my part i think "ambition, not to mention audacity" has a nice ring to it.

thanks for noticing the project. i am not looking for 'true believers', i am not asking people to set their skeptisim aside; i want people to notice, to ask questions. its up to us to provide the answers and to build this thing. getting people to pay attention is the first critical step, so, for your help with that, thanks.

we are a long ways from 'bolting this thing together..." as others have said, but i do think we are on the right track. its not gonna be easy, but it will work.

take care. mjl


Anyone here ever take differential equations ? The beginning of a 'S' curve, or logistic function, will be approximately linear. To note that now its incremental advances and declare thats how it will continue to be is a premature assumption. I'd hate to see what your timeframe on the development of PCs would have been if you had made such predictions in the 1950's about the potential of transistor technology.

However, an equally premature assumption is that the proliferation of human controlled molecular manipulation will follow the same sort of double exponential curve that Moore's law claims. Actually, i think thats a bad model even for computers and advocate a logistic S curve myself, but thats not my point.

To say the 21st century is not going to be a wild ride is a pessimistic and reactionary based on the possibility of beginning an S curve, while claims of assemblers by 2015 are absurd and defy common sense.


The 1st couple chapters of "Age of Spiritual Machines" deliniated everything I've read about reflexivity and non-linearity in economic systems, into (potentially)concrete technological and engineering applications. Even if only computers follow this curve they can enable a whole bunch of other technologies on the way.


I agree with that, cdnprodigy. Thats why i take a middle ground with this. I thought the remark "The 21st century will NOT be a wild ride" was quite reactionary and a premature conclusion. I think we'll see some amazing stuff in our lifetime, but may have to wait quite some time on Drexlerian machines. Predicting any further out than 20 yrs is futile, so i'll leave this conversation saying i don't know when but longer than 20 years for certain for an assembler.

I do dispute the fact that the nanotech industry is exactly where it was in 1990. No one had ever made a 'self assembled' circuit then, mass produced nanotubes, etc. However crude these improvements are, as i mentioned, the beginning of a logistic curve. Directed self assembly is the first step to understanding how to make an assembler, and is something we've made inroads on since 1990.


In fact, we are heading towards the end of the S-curve.

Mike Treder and Chris Phoenix are seriously naive and deluded. They are looking at the dubious "Law of Accelerating Returns" for reference.


MMG, are you really Phillip?

Chris Phoenix, CRN

Joe, in predicting exponential and then integrated molecular manufacturing, I don't rely on double exponential analysis. Instead, I look at three things:

1) How close are we to being able to achieve a very basic set of nanoscale fabrication tools that are precise and programmable?

2) Given a programmable molecular manufacturing system, even a primitive one, how much easier will that make it to build better systems?

3) How much incentive will there be (for at least a few large entities) to throw billions of dollars at the project, once it's realized that it's achievable?



My name is Michael, not Phillip.

Philip Moriarty

I can also confirm that 'MysticMonkeyGuru' is not a pseudonym for Philip Moriarty! (Though I can see why Unknown made the suggestion...)

Best wishes,



Unknown, are you a hot blonde women or Hannibal Lector? People with friendly intentions generally don't leave a reply address to "nowhere"...


We're headed toward the end of the S curve in traditional microelectronics for sure. Not to say computing power definitely won't increase, but a new paradigm will be needed for that to continue. Outside of current microelectronics technology, i'm not sure with respect to what you mean we're nearing the end of the S curve.

I don't see how you can say we're near the end of the S curve in nanotechnology research (if thats what is being referred to) if we haven't even established such a model is appropriate to be applied to advances in that field.

Tom Craver

I hate to do it, but I'm going to weigh in supporting what MMGuru is saying - to a degree. In some areas, science/technology does appear to be running slower than it should. Particularly medicine, energy, transportation. Conventionally defined nanotech appears to be making fairly good progess right now, but that could still bog down or become overly specialized.

Why? Corporations generally won't invest in much they can't quickly commercialize. Government-grant focused academic researchers get sucked into a "process" orientation that rewards "studies" and "investigations" rather than focusing on achieving goals leading to cures or other breakthroughs.

The right solution, for which I'm hearing a slowly increasing chorus of voices, is achievement prizes. Do something worthwhile, get a big prize to continue your work, and of course substantial ego gratification. Others have noted that many times groups will spend more to win a prize, than the prize itself is worth.

CRN is focused on "safe onset of MNT", but they have said that an earlier, and highly visible onset is likely to be safer than a later (hence faster) and/or secretive development. How about creating a list of achievements that are "prize-worthy", even if it falls to someone else to arrange for and finance the prizes?


Tom, the prizes sound like a great idea. And you are right, corporate investments will play a role in the speed of the development of this technology. Intel is beginning to get in, IBM has been in, HP has been in. None of them seem to be 'Manhattan Project' type efforts though. I'm not well briefed on how much they're spending, but it is a step even if its small.

I think 20+ years is a fair thing to say for the full maturation of this technology (gives me a lot of wiggle room too !!). I don't think it will be 50 years, but as i said above .. predicting beyond 20 yrs is futile.

Brett Bellmore

"In some areas, science/technology does appear to be running slower than it should. Particularly medicine, energy, transportation."

Essentially, areas which are highly regulated. Our ideas of how fast technology "ought" to progress are largely based on industries which were, during their growth periods, left pretty much alone. Electronics. Software. Early aviation. They faced no artificial barriers to advancement.

Today we face the regulatory state, where the presumption is that if people are going to do something, it's going to be regulated. Why, so complete is the victory of the regulatory mindset, that I've even seen advocates for advanced technology dedicate substantial brainpower that ought to go to pushing the technology forward, dreaming up complex regulatory schemes to be imposed before the technology even arrives.


Indeed. The innovations of the past century, aviation, information technology, medicine, electronics and countless more proceeded quickly without heavy regulation or government interference. Today, it's a different story. Most 1950s visions of the future, flying cars, Moon and Mars colonies, food pills, didn't materialize properly, not just because of the technological limitations of the era, but also because of regulations and safety issues. The same thing will happen in regards to nanotechnology and advanced AI and medicine. Technologically we are not as nearly as advanced as what Mike Treder claims on this site and on "Incipient Posthuman".

Chris Phoenix, CRN

MMG, you seem to be making two different arguments. One, that we are not advanced. Two, that regulation has slowed us down and will keep doing so.

Today, we can machine structures that are smaller than the molecules we can build. I'd call that pretty advanced. And I think we are continuing to advance pretty fast in the lab, though the consumer may not see it. It used to be that twenty years was half a career. Now, it's almost ancient. Imagine someone who's been out of a scientific field for twenty years trying to get a job in that field. Even after ten years, they'd need a lot of retraining. (I say this with some discomfort, having left the software field eight years ago.)

A question: Do you expect regulation to slow things down worldwide? Or do you expect that progress will be rapid elsewhere? Or is there a third possibility?


Tom Craver

Chris -

Can you point to any reasonably developed nation that funds science but does NOT heavily regulate the process? From the "Spiked" survey, this seemed relevant to this discussion of govt impact on research:


It seems to me that we waste much or most govt research money on non-goal-driven research, especially in areas where there clearly ARE important goals worth attaining, that justified massive funding in the first place.

The critical difference between successful govt programs and ineffective programs likely lies in a distinction between regulating and policing.

Regulation tries micro-monitor and micro-manage to suppress fraud and keep funds away from incompetent (or at least un-credentialed) researchers. But in so doing, it creates a process that makes the whole program wasteful and ineffective - the very things the process was intended to prevent.

A "policing" approach would not try to directly stop fraud or incompetence. Rather it would create a credible risk that such will be detected and VERY stingently punished - fines, possibly jail, certainly ruined reputation. Funding would be distributed in some expeditious fashion - likely assigning funding to research organizations with the firmly stated expectation that they will see to it the money is well spent - or else.

There'd certainly be inefficiencies - department politics, etc. But it'd still end up far more productive, as those with money to locally allocate will be better positioned to understand the impact of proposed lines of research and the effectiveness of individual researchers. At all levels, there'd be a direct incentive to push for effectiveness and avoid fraud.

The comments to this entry are closed.