Contacts: Chris Phoenix (1-305-387-5583) cphoe...@CRNano.org Director of Research Mike Treder (1-718-398-7272) mtre...@CRNano.org Executive Director
Nanobots Not Needed
SUMMARY: The popular idea of so-called nanobots, powerful and at risk of running wild, is not part of modern plans for building things "atom-by-atom" by molecular manufacturing. Studies indicate that most people don't know the difference between molecular manufacturing, nanoscale technology, and nanobots. Confusion about terms, fueled by science fiction, has distorted the truth about advanced nanotechnology. Nanobots are not needed for manufacturing, but continued misunderstanding may hinder research into highly beneficial technologies and discussion of the real dangers.
Nanobots have plagued nanotechnology from the beginning. Eric Drexler's "Engines of Creation" (1986), which introduced nanotechnology to the public, described certain kinds of tiny robots with limited capability. But in some fiction and fanciful speculation, these "nanorobots" or "nanobots" possess near-magical powers: transforming any object into anything else, acting as a universal medical device, or destroying anything they touch. This idea has caused confusion about the actual goals of advanced nanotechnology[1] research.
Originally, nanotechnology was about building stuff from the atoms up. "Assemblers" were specialized molecular construction machines. "Disassemblers" were research tools to figure out how to make things. A programmable atom-based manufacturing system would be able to build as many more systems as desired. But all these ideas merged with the nanobot concept, plus a heavy dose of science fiction, to create the idea of a single machine that could do it all-and might run wild, turning the world into a "gray goo" of self-copies.[2]
Meanwhile, the meaning of "nanotechnology" was being stretched. As funding opportunities increased, researchers in related and distant fields of nanoscale technology adopted the term to describe work they'd been doing for decades. By 1992, Drexler had to coin "molecular manufacturing" and "molecular nanotechnology" to indicate what he originally meant by nanotechnology.
Studies have shown that most readers don't know the difference between molecular manufacturing, nanoscale technology, and nanobots. Most nanoscale technologies use big machines to make small products. Molecular manufacturing is about tiny manufacturing systems. But those manufacturing systems are not nanobots.[3] Modern plans for molecular manufacturing do not involve self-contained nanoscale construction robots at all.
No one worries about an inkjet printer crawling off the desk and stealing ink cartridges. Molecular manufacturing systems will be no more autonomous than inkjets. Early, primitive, microscopic systems will not even have onboard computers. In advanced designs, called nanofactories,[4] the molecular fabrication apparatus will all be fastened down in well-ordered ranks inside a much larger structure. All designs will be externally controlled and supplied, capable of producing a duplicate nanofactory in about an hour-but only on command.
As nanoscale technologies begin to move from the lab to the marketplace, and attention turns to molecular manufacturing research, it will be increasingly important for journalists to counter outdated and incorrect ideas of nanotechnology and molecular manufacturing. Both scientists and the public have gotten the idea that molecular manufacturing requires the use of nanobots, and they may criticize or fear it on that basis. The truth is less sensational, but its implications[5] are equally compelling.
The Center for Responsible Nanotechnology is headquartered in New York. CRN is an affiliate of World Care, an international, non-profit, 501(c)(3) organization.
NOTES:
[1] Nanotechnology has several definitions. Today, a widely accepted definition is any technology involving structures between 1 and 100 nanometers with novel properties. (A nanometer is a billionth of a meter, roughly the length your fingernails grow in one second.) There are many ways of building nanoscale structures and materials, and for each there is a different branch of nanotechnology. Most of these nanoscale technologies use large tools to create small structures. In general, these can be understood as traditional industrial or chemical processes, and not the same thing as molecular manufacturing. For more, see "What is Nanotechnology?" at http://www.crnano.org/whatis.htm, and "What Is Molecular Manufacturing?" at http://www.crnano.org/essays05.htm#2,Feb
[3] In the long term, some products of molecular manufacturing systems could be nanobots (e.g., for medical use), but these are not envisioned to be metabolizing or self-replicating (at least not by credible researchers).
[5] For information on the risks and benefits of advanced nanotechnology, see "CRN Research: Overview of Current Findings" at http://www.crnano.org/overview.htm
In article <d08gdc01...@enews4.newsguy.com>, Mike Treder, CRN wrote:
Um. One hates to come off sounding rude or snarky, but... is this the whole document?
> Nanobots Not Needed > SUMMARY: The popular idea of so-called nanobots, powerful and at risk > of running wild, is not part of modern plans for building things > "atom-by-atom" by molecular manufacturing. Studies indicate that most > people don't know the difference between molecular manufacturing, > nanoscale technology, and nanobots. Confusion about terms, fueled by > science fiction, has distorted the truth about advanced nanotechnology. > Nanobots are not needed for manufacturing, but continued > misunderstanding may hinder research into highly beneficial > technologies and discussion of the real dangers.
I will accept the assertion that most people do not, in fact, know the difference between those terms. However, a footnote to the studies referenced would make this rather more persuasive. It would make it more than an assertion, anyway. I also claim that another factor in the confusion is the tendency to up and change names every few years.
More importantly than either of those two criticisms, is the final assertion-- that this confusion somehow might hinder research. There's a clever shift of the pen there from (explicitly) "most people" to (implicitly) people doing actual research. I'm pretty sure that the researchers and the layer or two of pocketbook holders are a little more clued in on those terms than the average everyman on the street. Very few people write seven or eight digit checks on a whim.
> Nanobots have plagued nanotechnology from the beginning. Eric Drexler's > "Engines of Creation" (1986), which introduced nanotechnology to the > public, described certain kinds of tiny robots with limited capability. > But in some fiction and fanciful speculation, these "nanorobots" or > "nanobots" possess near-magical powers: transforming any object into > anything else, acting as a universal medical device, or destroying > anything they touch. This idea has caused confusion about the actual > goals of advanced nanotechnology[1] research.
This, too, would have been an excellent place to insert a footnote about studies supporting your claims. In this case, a study supporting the claim of layman confusion about the terms would again have done a world of good. So would a good scientific survey of just exactly what "advanced nanotechnology research" actually is.
Instead, what we get is a footnote with boilerplate about nanotechnology and millionths of meters, without any support for the claims being made. What this does, in effect, is presumptively-- even presumptuously-- define everyone working on something other than this paper's preferred course of research to be something other than the actual goal of advanced nano.
I think that's a bit of an overstretch. Don't you? Unless, uh, you actually are speaking for all the nanotech researchers out there.
> Originally, nanotechnology was about building stuff from the atoms up. > "Assemblers" were specialized molecular construction machines. > "Disassemblers" were research tools to figure out how to make things. A > programmable atom-based manufacturing system would be able to build as > many more systems as desired. But all these ideas merged with the > nanobot concept, plus a heavy dose of science fiction, to create the > idea of a single machine that could do it all-and might run wild, > turning the world into a "gray goo" of self-copies.[2]
> Meanwhile, the meaning of "nanotechnology" was being stretched. As > funding opportunities increased, researchers in related and distant > fields of nanoscale technology adopted the term to describe work they'd > been doing for decades. By 1992, Drexler had to coin "molecular > manufacturing" and "molecular nanotechnology" to indicate what he > originally meant by nanotechnology.
Yes, and then... what was it? Zettatechnology? I submit that this solves nothing, and contributes to the nomenclature confusion.
> Studies have shown that most readers don't know the difference between > molecular manufacturing, nanoscale technology, and nanobots. Most > nanoscale technologies use big machines to make small products. > Molecular manufacturing is about tiny manufacturing systems. But those > manufacturing systems are not nanobots.[3] Modern plans for molecular > manufacturing do not involve self-contained nanoscale construction > robots at all.
Again, cite the studies. If there's anything that gets up my nose, it's unsubstantiated claims which are repeated like sledgehammer blows. What studies? Cite them. How were they conducted? Who conducted them? You are dedicating an entire essay to the refutation of these claims-- the reader deserves the opportunity to see them.
(Also, as a sidenote, it is customary to footnote someone other than your own organization. At least once.)
Also, be very, very careful about absolute claims. Your footnote here is useful, in that it prevented me from bringing up the numerous possible uses for nanobots (e.g., medical, which you note; micromaintenance; sensing and monitoring; and other infrastructure uses.) But is also contained the absolute statement that *no* credible researchers are planning nanobots where are either metabolising or self-reproducing.
And yet, I am presently reading a book chapter (found online at: http://www.bionano.neu.edu/Nanorobotics.pdf ) which very clearly outlines their desire to develop nanobots capable of self-replication when required. They also specifically express the desire to develop a toolkit of standard parts including rhodopsin, bacteriorhodopsin, and artificial structures derived from those, to act as solar collectors for on-site power gathering. If this is not metabolism as you conceive it, I'm not sure what does. So I ask, does solar collection match your definition of metabolism?
I must also ask if the authors (Ummat, Dubey, Sharma, and Mavroidis) are to be considered credible researchers, or not.
There is a similar paper by a subset of those authors (found online here: http://www.bionano.neu.edu/Bionanorobotics_Chapter%207_upload.pdf ) which I will admit I have not read (it's on my list after a paper on artificial bacterial foraging strategies, and an application of same to electrical design) which dedicates several pages to the topic of self-replicating nanobots, with a call for more research on the topic and a mention of ongoing research in the area.
In short, your paper is full of assertions, some of which are unsubstantianted, some of which I believe to be simply incorrect.
I have no further comments on the rest of your paper, but I include the rest of it for the convenience of other readers.
> No one worries about an inkjet printer crawling off the desk and > stealing ink cartridges. Molecular manufacturing systems will be no > more autonomous than inkjets. Early, primitive, microscopic systems > will not even have onboard computers. In advanced designs, called > nanofactories,[4] the molecular fabrication apparatus will all be > fastened down in well-ordered ranks inside a much larger structure. All > designs will be externally controlled and supplied, capable of > producing a duplicate nanofactory in about an hour-but only on > command.
> As nanoscale technologies begin to move from the lab to the > marketplace, and attention turns to molecular manufacturing research, > it will be increasingly important for journalists to counter outdated > and incorrect ideas of nanotechnology and molecular manufacturing. Both > scientists and the public have gotten the idea that molecular > manufacturing requires the use of nanobots, and they may criticize or > fear it on that basis. The truth is less sensational, but its > implications[5] are equally compelling.
> The Center for Responsible Nanotechnology is headquartered in New York. > CRN is an affiliate of World Care, an international, non-profit, > 501(c)(3) organization.
> NOTES:
> [1] Nanotechnology has several definitions. Today, a widely accepted > definition is any technology involving structures between 1 and 100 > nanometers with novel properties. (A nanometer is a billionth of a > meter, roughly the length your fingernails grow in one second.) There > are many ways of building nanoscale structures and materials, and for > each there is a different branch of nanotechnology. Most of these > nanoscale technologies use large tools to create small structures. In > general, these can be understood as traditional industrial or chemical > processes, and not the same thing as molecular manufacturing. For more, > see "What is Nanotechnology?" at http://www.crnano.org/whatis.htm, and > "What Is Molecular Manufacturing?" at > http://www.crnano.org/essays05.htm#2,Feb
> [3] In the long term, some products of molecular manufacturing systems > could be nanobots (e.g., for medical use), but these are not envisioned > to be metabolizing or self-replicating (at least not by credible > researchers).
> [5] For information on the risks and benefits of advanced > nanotechnology, see "CRN Research: Overview of Current Findings" at > http://www.crnano.org/overview.htm
-- John S. Novak, III j...@cegt201.bradley.edu The Humblest Man on the Net
> In article <d08gdc01...@enews4.newsguy.com>, Mike Treder, CRN wrote:
> Um. One hates to come off sounding rude or snarky, but... is this the > whole document?
I'm afraid I had a reaction similar to John's. I agree with all his comments and the best I can do is rephrase some of his objections and throw in some of my own observations and opinionated opinions:
>> Nanobots Not Needed
>> SUMMARY: The popular idea of so-called nanobots, powerful and at risk >> of running wild, is not part of modern plans for building things >> "atom-by-atom" by molecular manufacturing.
John's already cited research efforts that void the claim that nanobots aren't part of "modern plans". I'll just add that *I* still consider "nanobots" part of the plan. The only aspect of the CRN release that I can agree with is the implication that the "risk of running wild" is what drives some of the fear of nanotechnology. But, as John notes, what matters is who is afraid.
>> Nanobots are not needed for manufacturing,
Nanotechnology of *any sort* isn't needed for manufacturing. Neither is a knowledge of quantum mechanics, since the industrial revolution was born prior to either one of these things. It isn't until you get specific about the problems you want to address if you had control at the molecular level that the need for nanobots can be addressed. Manufacturing, per se, is probably a Red Herring when it comes to addressing the benefits and risks of nanotechnology.
>> Nanobots have plagued nanotechnology from the beginning.
I wonder, was the double entendre in that sentence intended or accidental? ;-)
> I have no further comments on the rest of your paper, but I include > the rest of it for the convenience of other readers.
Thanks - I have a few more comments.
>> No one worries about an inkjet printer crawling off the desk and >> stealing ink cartridges. Molecular manufacturing systems will be no >> more autonomous than inkjets. Early, primitive, microscopic systems >> will not even have onboard computers. In advanced designs, called >> nanofactories,[4] the molecular fabrication apparatus will all be >> fastened down in well-ordered ranks inside a much larger structure. >> All designs will be externally controlled and supplied, capable of >> producing a duplicate nanofactory in about an hour-but only on >> command.
Comparison with a relatively benign machine, an inkjet printer, belies the large difference in potential capabilities and impact. On the one hand we are to believe that a molecular manufacturing system would not have sufficient intelligence to run autonomously, yet on the other hand it might need to be intelligent enough to prohibit the manufacture of anything dangerous. If there are no limits on what a nanofactory can produce, it can hardly be compared to an inkjet printer - the differences in output capabilities are humongous. But the intelligence that would need to be incorporated in such a device to limit it to benign output would, in my humble opinion, definitely accord it the capability of crawling off the table to steal raw material!
The ink-jet comparison also happens to highlight an economic problem with the nanofactory approach: just as ink-jet cartridges are a large profit center for printer manufacturers, any use of preprocessed raw material for nanofactories creates a supply choke-hold on users. The economic miracle that would occur from the use of self-replicating devices that are free from any form of central control (e.g. able to use unprocessed raw material from their local environment) is cruelly denied.
In my humble opinion, nanotechnology's greatest benefits are in improvements in medicine, the alleviation of poverty, and increase in individual freedoms (due to possible decoupling from the global economy). I believe the primary motivation of promoting nanofactories over nanobots is to somehow mitigate the risks of nanotechnology. I consider such risk avoidance dangerously misguided since, if successful, it would hinder and delay the really meaningful benefits that only nanotechnology could provide.
In article <112imdp3h89d...@news.supernews.com>, Jim Logajan wrote: >>> Nanobots are not needed for manufacturing, > Nanotechnology of *any sort* isn't needed for manufacturing. Neither is a > knowledge of quantum mechanics, since the industrial revolution was born > prior to either one of these things. It isn't until you get specific > about the problems you want to address if you had control at the > molecular level that the need for nanobots can be addressed. > Manufacturing, per se, is probably a Red Herring when it comes to > addressing the benefits and risks of nanotechnology.
I would add that part of the specificity is an economic question. Many, many things that can be manufactured, can be manufactured "better" or "more economically" with molecular assembly. (Or so it is claimed, and I'm not dispputing that.)
>>> No one worries about an inkjet printer crawling off the desk and >>> stealing ink cartridges. Molecular manufacturing systems will be no >>> more autonomous than inkjets. Early, primitive, microscopic systems >>> will not even have onboard computers. In advanced designs, called >>> nanofactories,[4] the molecular fabrication apparatus will all be >>> fastened down in well-ordered ranks inside a much larger structure. >>> All designs will be externally controlled and supplied, capable of >>> producing a duplicate nanofactory in about an hour-but only on >>> command.
> Comparison with a relatively benign machine, an inkjet printer, belies > the large difference in potential capabilities and impact. On the one > hand we are to believe that a molecular manufacturing system would not > have sufficient intelligence to run autonomously, yet on the other hand > it might need to be intelligent enough to prohibit the manufacture of > anything dangerous. If there are no limits on what a nanofactory can > produce, it can hardly be compared to an inkjet printer - the differences > in output capabilities are humongous. But the intelligence that would > need to be incorporated in such a device to limit it to benign output > would, in my humble opinion, definitely accord it the capability of > crawling off the table to steal raw material!
This is an important point, to which no one has *the* answer. If, fifty years from now, a basement workshop appliance can rip out pneumonic plague spores, then we're going to have problems-- especially when one considers the intersection of benchtop nanoconstruction with the internet and encryption techniques. Gives a whole new meaning to the term "computer virus."
Ah, but of course, that sort of thing poses a special problem for this essay, doesn't it? By a reasonably broad interpretation, plague spores are self-replicating, metabolizing nanobots... aren't they? If they're not, I invite the CRN people to tell me why not, but that's really a side issue. The real issue I'm raising here is that, even though the CRN people are waving their arms telling me that self-replicating, self-metabolizing nanobots aren't necessary and are a bad idea *doesn't mean that everyone else will agree*.
And becuase not everyone else will agree, that makes it increasingly likely as time goes on that they will eventually be developed. CRN telling me airily that it's not necessary doesn't mean that it won't happen, and in my mind CRN seems to be adopting (intentionally or not) a sort of a Bill Joy-esque Reliquishment type of argument. "Well, uh, let's not do that."
My response remains: "Then we will be totally unprepared when someone unscrupulous does exactly that."
> The ink-jet comparison also happens to highlight an economic problem with > the nanofactory approach: just as ink-jet cartridges are a large profit > center for printer manufacturers, any use of preprocessed raw material > for nanofactories creates a supply choke-hold on users. The economic > miracle that would occur from the use of self-replicating devices that > are free from any form of central control (e.g. able to use unprocessed > raw material from their local environment) is cruelly denied. > In my humble opinion, nanotechnology's greatest benefits are in > improvements in medicine, the alleviation of poverty, and increase in > individual freedoms (due to possible decoupling from the global economy).
I am obligated to snort and chuckle. Resources are not infinitie, even with molecular assembly. I see no decoupling from the global economy in the near future, nor do I see it as desireable, unless one uses the technology to leave the globe.
> I believe the primary motivation of promoting nanofactories over nanobots > is to somehow mitigate the risks of nanotechnology. I consider such risk > avoidance dangerously misguided since, if successful, it would hinder and > delay the really meaningful benefits that only nanotechnology could > provide.
And leave us flatfooted when the dangerous stuff does, inevitably, come along.
-- John S. Novak, III j...@cegt201.bradley.edu The Humblest Man on the Net
> In article <112imdp3h89d...@news.supernews.com>, Jim Logajan wrote: >> In my humble opinion, nanotechnology's greatest benefits are in >> improvements in medicine, the alleviation of poverty, and increase in >> individual freedoms (due to possible decoupling from the global >> economy).
> I am obligated to snort and chuckle.
Happy to have provided you a bit of mirth and merriment. ;-)
> Resources are not infinitie, even with molecular assembly.
Nanotechnology should make more resources accessible while at the same time making the need for them, presumably, less.
> I see no decoupling from the global economy in the near future,
Well... I don't know how many pre-industrial tribes still exist, but I seem to recall there were still quite a few extant not that many decades ago. I believe they were effectively "decoupled" from the global economy. Except for omni-present government presence, even today one can manage a modest amount of decoupling, depending on the sacrifices one is willing to make. That said, I don't see any mass outbreak of neo-rugged individualists armed with nanobots making a break from the global industrial economy anytime soon either.
> nor do I see it as desireable, unless one uses the technology > to leave the globe.
The capability to decouple from the global economy would seem to be an essential requirement for any sustained expansion into space, but I'm not sure why you think any reduction of interaction with the global economy is undesirable for those who remain on Earth. Obviously some aspects cannot be decoupled (e.g. usage of real-estate) but certainly much economic interaction can be reduced. Are you, perhaps, concerned about the onset of a "tragedy of the commons" more severe than that which exists now?
> I am obligated to snort and chuckle. Resources are not infinitie, > even with molecular assembly. I see no decoupling from the global > economy in the near future, nor do I see it as desireable, unless one > uses the technology to leave the globe.
.... Wherever the resources for self-sufficient farming are available, the resources for self-sustaining nanotech that doesn't need a gardener, tools that can't be made onsite, nitrogen salts, or much water are also available, the only ongoing inputs from the global 'economy' being new information where upgrades are desired.
> In article <d08gdc01...@enews4.newsguy.com>, Mike Treder, CRN wrote:
> Um. One hates to come off sounding rude or snarky, but... is this the > whole document?
Yes, it's the whole document.
You do come off sounding rude and snarky. Let's see if I can answer without sounding defensive.
> I will accept the assertion that most people do not, in fact, know the > difference between those terms. However, a footnote to the studies > referenced would make this rather more persuasive.
I was thinking of two general nanotech studies, one in the US and one in Britain, showing that many people still have never heard of nanotech, and many of those who have can't give any kind of definition. I don't have the references handy, but they've been in the news.
> More importantly than either of those two criticisms, is the final > assertion-- that this confusion somehow might hinder research. There's > a clever shift of the pen there from (explicitly) "most people" to > (implicitly) people doing actual research. I'm pretty sure that the > researchers and the layer or two of pocketbook holders are a little > more clued in on those terms than the average everyman on the street. > Very few people write seven or eight digit checks on a whim.
I was not saying that researchers might be confused, but that popular confusiong might do political damage that could hurt research.
I'm not so sure that funders aren't confused. Some of them don't know what molecular manufacturing is about--they just know that they're opposed to Drexler.
In any case, there was no shift of the pen.
> .... So would a good scientific survey of just > exactly what "advanced nanotechnology research" actually is.
> Instead, what we get is a footnote with boilerplate about > nanotechnology and millionths of meters, without any support for the > claims being made. What this does, in effect, is presumptively-- even > presumptuously-- define everyone working on something other than this > paper's preferred course of research to be something other than the > actual goal of advanced nano.
Well, from one point of view, everyone doing scientific research into nanotech is doing "advanced" work. That's not how I meant it. I think a reasonable definition for advanced nanotech (in the absence of any formal definition for a non-technical phrase) is the ability to build precise molecular structures incorporating terabytes of blueprints. I don't see any way to do that other than molecular manufacturing. This is going out on a limb a bit, so feel free to suggest technical alternatives.
>>Studies have shown that most readers don't know the difference between >>molecular manufacturing, nanoscale technology, and nanobots. Most >>nanoscale technologies use big machines to make small products. >>Molecular manufacturing is about tiny manufacturing systems. But those >>manufacturing systems are not nanobots.[3] Modern plans for molecular >>manufacturing do not involve self-contained nanoscale construction >>robots at all.
> Again, cite the studies. If there's anything that gets up my nose, > it's unsubstantiated claims which are repeated like sledgehammer > blows. What studies? Cite them. How were they conducted? Who > conducted them? You are dedicating an entire essay to the refutation > of these claims-- the reader deserves the opportunity to see them.
Sorry it got up your nose. In the original version, the formatting made it clear that the first paragraph was a summary of the rest of the document, so the claim was not repeated, just restated.
It was not written as an academic paper, and I tried to avoid dull phrasing; if you were expecting academic work with footnotes, then you may have seen the style as "sledgehammer blows." Sorry.
> Also, be very, very careful about absolute claims. Your footnote here > is useful, in that it prevented me from bringing up the numerous > possible uses for nanobots (e.g., medical, which you note; > micromaintenance; sensing and monitoring; and other infrastructure > uses.) But is also contained the absolute statement that *no* > credible researchers are planning nanobots where are either > metabolising or self-reproducing.
> And yet, I am presently reading a book chapter (found online at: > http://www.bionano.neu.edu/Nanorobotics.pdf ) which very clearly > outlines their desire to develop nanobots capable of self-replication > when required.
This chapter certainly includes uncautious language. I'm surprised it didn't get up your nose as well: "This programming capability would form the core essence of a bionano robotics system and hence enables them with immense power." (p. 23) And their illustration 15A (p. 22) is only an artist's conception--the protein helices are smaller than atoms, and it includes a propeller for motility--but it's described as though it were an actual proposal.
> They also specifically express the desire to develop a > toolkit of standard parts including rhodopsin, bacteriorhodopsin, and > artificial structures derived from those, to act as solar collectors > for on-site power gathering. If this is not metabolism as you > conceive it, I'm not sure what does. So I ask, does solar collection > match your definition of metabolism?
Toolkits are good. I liked what they said about developing toolkits in the first part of the chapter.
No, solar collection does not fit what I mean by metabolism. Metabolism requires at least the ability to arrange (e.g. break down) disordered chemicals into a more useful state. Solar collection is energy supply. An electric motor does not metabolize.
> I must also ask if the authors (Ummat, Dubey, Sharma, and Mavroidis) > are to be considered credible researchers, or not.
I don't know the researchers, though I recognize Mavroidis's name. But the quality of the chapter is inconsistent. Parts are, frankly, semi-literate. "This is a traditional method, which has been in use since quite sometime for designing bio molecules." This may be a second-language problem, but it should have been edited. I also note that the first reference gets Drexler's name in the wrong order.
Other parts of the chapter are over-ambitious and under-supported. Without knowing who wrote which part (and suspecting that some parts may have been written/compiled by anonymous grad students) I would hesitate to task the listed authors with the wilder projections.
But perhaps you're right--it appears possible that these researchers are actually proposing free-floating self-contained self-actuated self-controlled molecular manufacturing systems. It's certainly possible to get that idea from reading the less technical parts of the chapter. I wouldn't have expected this. Maybe I'll write Mavroidis and see whether he intended to give that impression.
It's great to see ideas of engineered molecular robotic systems get published. But I wish they had not been published in a chapter that listed evolution as a desirable property of nanosystems, with no support or clarification.
Note that in Figure 18, the machines are fastened down. That doesn't fit the popular conception of nanobots.
The long-term projections and descriptive rhetoric in this chapter do appear to contradict what I wrote. The actual technology and plans appear far more sedate. In the end, I'm left unsure whether they're planning externally controlled, fastened-down, engineered machines and only mentioning the other stuff to be futuristic, or whether they actually think evolution and autonomy are desirable research goals.
> There is a similar paper by a subset of those authors (found online here: > http://www.bionano.neu.edu/Bionanorobotics_Chapter%207_upload.pdf ) > which I will admit I have not read (it's on my list after a paper on > artificial bacterial foraging strategies, and an application of same > to electrical design) which dedicates several pages to the topic of > self-replicating nanobots, with a call for more research on the topic > and a mention of ongoing research in the area.
The paper has a lot of overlap with the previously cited one. (The English is better.) I note that their approach to self-replication appears to be limited to a self-templating self-assembly paradigm: "To create any system with self replicating mechanism we need to first find out its most stable state, then we need to calculate its behavior in the extrinsic gradients and then we need to excite it with energy and supply of intrinsic materials so that it replicates."
Everyone, please copy me by email on any reply you want me to see; I don't usually have time to follow sci.nanotech discussions.
Chris
-- Chris Phoenix cphoe...@CRNano.org Director of Research Center for Responsible Nanotechnology http://CRNano.org
Mike Treder, CRN wrote: > SUMMARY: The popular idea of so-called nanobots, powerful and at risk > of running wild, is not part of modern plans for building things > "atom-by-atom" by molecular manufacturing.
Unfortunately this summary also summarises what is wrong with this document. There's this story where two people discuss and it goes like this:
Person A: No real 68-er eats fish-fingers. Person B: But, person so-and-so *DOES* eat fish-fingers. Person A: Yes, but no *REAL* 68-er eats fish-fingers.
Your blanket statement that nanobots are not part of modern plans for building things atom-by-atom is similar. There are indeed plans for building things atom-by-atom with the help of nanobots, but you would presumably anser: Yes, but no *MODERN* plans include nanobots.
> Studies indicate that most people don't know the difference between > molecular manufacturing, nanoscale technology, and nanobots.
And here is the second problem;
*WHICH* studies ? No, that doesn't need to be in the introduction, but it should be SOMEWHERE. You make a lot of rather strong claims in this document, and say that several of them are supported by studies. But not a single time do you reference any of those studies.
> Nanobots have plagued nanotechnology from the beginning. Eric > Drexler's "Engines of Creation" (1986), which introduced > nanotechnology to the public, described certain kinds of tiny robots > with limited capability. But in some fiction and fanciful speculation, > these "nanorobots" or "nanobots" possess near-magical powers: > transforming any object into anything else, acting as a universal > medical device, or destroying anything they touch. This idea has > caused confusion about the actual goals of advanced nanotechnology[1] > research.
This I agree with. It's even quite understandable. A cursory reading of Engines of Creation *does* indeed give you the impression the things are nearly magical. You need to spend quite a bit of time before you start discovering that there *are* limits, and what those limits are.
Not all of the devices described by Drexler are all that limited by the way. He includes the general assembler. A general assembler as described by drexler certainly could not in any way turn into grey goo, but it's also not exactly a device with "limited capability".
> Studies have shown that most readers don't know the difference between > molecular manufacturing, nanoscale technology, and nanobots.
Again: Which studies ?
By the way, this is likely true for *any* modern technology.
> Molecular manufacturing is about tiny manufacturing systems. But those > manufacturing systems are not nanobots.
More unsubstantiated claims. Why not ? For that matter, since you're obviously trying to clear up misunderstandings, how come that nowhere in the paper do you explain what exactly, in your opinion, a nanobot is. You give examples of things that in your opninion are *not* nanobots, but you fail to say why, and you fail to say what a nanobot is. This doesn't clear up anything at all.
> No one worries about an inkjet printer crawling off the desk and > stealing ink cartridges.
An inkjet printer is also not capable of producing a new inkjet-printer.
> Molecular manufacturing systems will be no more autonomous than > inkjets.
In article <112rt9cjp5ep...@news.supernews.com>, Chris Phoenix wrote: >> Um. One hates to come off sounding rude or snarky, but... is this the >> whole document? > Yes, it's the whole document. > You do come off sounding rude and snarky. Let's see if I can answer > without sounding defensive.
Having so recently chastened John Larkin for attacking a press release rather than the actual article under consideration, I had no wish to make the same mistake. I had to ask.
>> I will accept the assertion that most people do not, in fact, know the >> difference between those terms. However, a footnote to the studies >> referenced would make this rather more persuasive. > I was thinking of two general nanotech studies, one in the US and one in > Britain, showing that many people still have never heard of nanotech, > and many of those who have can't give any kind of definition. I don't > have the references handy, but they've been in the news.
That's, uh, not what I would consider a sufficient or a reasurring answer. A footnote is a pretty simple thing to add.
>> More importantly than either of those two criticisms, is the final >> assertion-- that this confusion somehow might hinder research. There's >> a clever shift of the pen there from (explicitly) "most people" to >> (implicitly) people doing actual research. I'm pretty sure that the >> researchers and the layer or two of pocketbook holders are a little >> more clued in on those terms than the average everyman on the street. >> Very few people write seven or eight digit checks on a whim. > I was not saying that researchers might be confused, but that popular > confusiong might do political damage that could hurt research.
I'm curious as to the mechanism you see for this damage.
> I'm not so sure that funders aren't confused. Some of them don't know > what molecular manufacturing is about--they just know that they're > opposed to Drexler.
I think most funders are smart enough to consult with experts (if they don't keep them on staff as a matter of course) before writing very large checks. Most big dollar funders are not fools.
Who is your intended audience for this piece?
>> Instead, what we get is a footnote with boilerplate about >> nanotechnology and millionths of meters, without any support for the >> claims being made. What this does, in effect, is presumptively-- even >> presumptuously-- define everyone working on something other than this >> paper's preferred course of research to be something other than the >> actual goal of advanced nano. > Well, from one point of view, everyone doing scientific research into > nanotech is doing "advanced" work.
That's sort of my point.
> That's not how I meant it. I think > a reasonable definition for advanced nanotech (in the absence of any > formal definition for a non-technical phrase) is the ability to build > precise molecular structures incorporating terabytes of blueprints. I > don't see any way to do that other than molecular manufacturing. This > is going out on a limb a bit, so feel free to suggest technical > alternatives.
This is still a "No True Scotsman" type of argument.
You could counter this with an argument from authority, but since this entire discussion (and I don't mean this Usenet thread, I mean this larger discussion in the field) turns on which authority you happen to believe, that course is not available to you.
The hard truth here is that there are multiple authorities, and they disagree with each other. The No True Scotsman approach, where you assert that only your definition of nanotechnology is advanced, is a classic fallacy. Even if I agreed with you, I would be morally obliged to point this fallacy out, and penalize you fifteen yards for unfair rhetoric.
My definition is not at issue. Yours-- and your support for it-- is.
>> Again, cite the studies. If there's anything that gets up my nose, >> it's unsubstantiated claims which are repeated like sledgehammer >> blows. What studies? Cite them. How were they conducted? Who >> conducted them? You are dedicating an entire essay to the refutation >> of these claims-- the reader deserves the opportunity to see them. > It was not written as an academic paper, and I tried to avoid dull > phrasing; if you were expecting academic work with footnotes, then you > may have seen the style as "sledgehammer blows." Sorry.
But you did give me a work with footnotes. You just didnt' give me a work with useful footnotes.
>> And yet, I am presently reading a book chapter (found online at: >> http://www.bionano.neu.edu/Nanorobotics.pdf ) which very clearly >> outlines their desire to develop nanobots capable of self-replication >> when required. > This chapter certainly includes uncautious language. I'm surprised it > didn't get up your nose as well: "This programming capability would form > the core essence of a bionano robotics system and hence enables them > with immense power." (p. 23) And their illustration 15A (p. 22) is only > an artist's conception--the protein helices are smaller than atoms, and > it includes a propeller for motility--but it's described as though it > were an actual proposal.
But I'm not critiquing that chapter, right now. I'm critiquing your article. There's really only one fundamental question to be answered, which got posed later on.
>> They also specifically express the desire to develop a >> toolkit of standard parts including rhodopsin, bacteriorhodopsin, and >> artificial structures derived from those, to act as solar collectors >> for on-site power gathering. If this is not metabolism as you >> conceive it, I'm not sure what does. So I ask, does solar collection >> match your definition of metabolism?
> Toolkits are good. I liked what they said about developing toolkits in > the first part of the chapter.
> No, solar collection does not fit what I mean by metabolism. Metabolism > requires at least the ability to arrange (e.g. break down) disordered > chemicals into a more useful state. Solar collection is energy supply. > An electric motor does not metabolize.
I could argue those semantics, but I don't need to; the ADP/ATP cycle is very much a part of their scheme, and seems to fit your definition as well.
>> I must also ask if the authors (Ummat, Dubey, Sharma, and Mavroidis) >> are to be considered credible researchers, or not.
And here's the fundamental question.
> I don't know the researchers, though I recognize Mavroidis's name. But > the quality of the chapter is inconsistent. Parts are, frankly, > semi-literate. "This is a traditional method, which has been in use > since quite sometime for designing bio molecules." This may be a > second-language problem, but it should have been edited.
I'm pretty sure that was a draft, but, reading a number of IEEE journals as part of my monthly diet has left me all but immune to the butchery of my native language.
Yes, even IEEE articles should be editted for grammar. Sadly, they aren't, and I am not in a position to reform the entire field. (I will, however, copy edit any technical paper anyone sends me for a nominal fee. I won't even ask for credit. And I'll spend more time worrying about it than I do for a typical Usenet post.)
> I also note > that the first reference gets Drexler's name in the wrong order.
> Other parts of the chapter are over-ambitious and under-supported. > Without knowing who wrote which part (and suspecting that some parts may > have been written/compiled by anonymous grad students) I would hesitate > to task the listed authors with the wilder projections.
> But perhaps you're right--it appears possible that these researchers are > actually proposing free-floating self-contained self-actuated > self-controlled molecular manufacturing systems. It's certainly > possible to get that idea from reading the less technical parts of the > chapter. I wouldn't have expected this. Maybe I'll write Mavroidis and > see whether he intended to give that impression.
> It's great to see ideas of engineered molecular robotic systems get > published. But I wish they had not been published in a chapter that > listed evolution as a desirable property of nanosystems, with no support > or clarification.
Yes, well. To be perfectly blunt about this, I could shorten some of my objections to your piece thus: "I wish you had not published a paper asserting public ignorance and defining contentious terms, with no support or clarification." (Well, you do footnote some other CRN papers, but that's just setting up CRN as the authority on the term, which I simply do not accept. I will not, in fact, accept any authority on these terms until someone by God builds something and becomes a real authority. I'm hardheaded that way. Those CRN papers don't even have authors listed on their primary pages.)
So here we are, then.
> Note that in Figure 18, the machines are fastened down. That doesn't > fit the popular conception of nanobots.
That is a diagram showing a nanobot assembly floor, e.g., a floor that is assembling nanobots. The figure does not imply that they remain on that floor while operating.
> The long-term projections and descriptive rhetoric in this chapter do > appear to contradict what I wrote. The actual technology and plans > appear far more sedate. In the end, I'm left unsure whether they're > planning externally controlled, fastened-down, engineered machines and > only mentioning the other stuff to be futuristic, or whether they > actually think evolution and autonomy are desirable research goals.
>> There is a similar paper by a subset of those authors (found online here: >> http://www.bionano.neu.edu/Bionanorobotics_Chapter%207_upload.pdf ) >> which I will admit I have not read (it's on my list after a paper on >> artificial bacterial foraging strategies, and an
