1) Put a cap on how many copies of itself a bot can make.
This does nothing to prevent a "gray goo" problem. All it takes is one "mutation"-- an error in manufacturing-- to remove this cap. In biological organisms, this is called "cancer," and it's something of a problem.
2) Add some sort of mechanism to the bots to only allow duplication when there are less than some number of bots in the surrounding area.
So rather than a hard limit on reproductions, you have an environmental feedback mechanism. Again, one mutation disables this feature, or worse, inverts it. Now the assembler only reproduces when the concentration in the local environment is greater than a given value.
3) Only allow bots to reproduce when told to do so by Humans.
Great idea. At this point, they're not self-assembling any more. This is, in fact, the only solution to the problem.
Of course, if the assemblers are able to self-assemble, it's quite likely that they will be caused to self-assemble by some external factor sooner or later. So the only safe way to apply this solution is to make the assemblers physically incapable of assembling themselves.
4) provide som sort of "master kill signal" (i.e. a specific radio frequency) that will cause all bots reciving it to cease functioning.
Just as it only takes one mutation-- or manufacturing error-- to disable a reproductive limit, so too does it take only one error to disable this idea. Too dangerous.
5) Any of several other ideas I didn't just think of in the past 2 min.
I hope the ones you think of in the next 2 minutes are somewhat better that these ones. For inspiration, look at living organisms, and observe all the things that can go wrong with them. Don't suggest anything that leads to one of those states.
And how do we know the goo would be grey, anyway?
The idea of a "gray goo" is favored by people who don't realize that nanotechnology is an old, old idea. There's already goo all over the place, but it isn't gray. It's green.
Re:Religious paranoid idiots will ban anything
on
Don't Stymie Nanotech
·
· Score: 5, Insightful
I'm stunned that four people felt compelled to give you mod points. Your comment is the most absurd one I've seen so far in this thread, and that's counting the trolls.
The wonderful thing about being a human being is that we can choose what to do and what not to do. I can choose to stop at a crosswalk, or I can choose to plow through a crowd of second-graders.
Your argument is basically that it's foolish to stop. Somebody out there is going to plow through a crosswalk anyway, so we might as well get in there and figure out how to use it to our advantage. If you think I'm misrepresenting your argument, maybe you'd better go back and read your own words again. "New technologies will come and be used whether we like it or not. Cloning will occur whether we ban it or not." You say this as if there were no moral or ethical aspect to it whatsoever, and that's simply not true.
Some things just should not be done. If you're an amoral person-- and your post, especially the part about "religious paranoid idiots," certainly seems to suggest that you are-- then you probably reject this assertion on its face. If that's the case, I won't bother trying to convince you otherwise. (My opinion is that people with no sense of morality or ethics at all are mentally ill in a way we just haven't figured out yet. Nothing personal; it's just a theory.)
So let's just take as given that you believe, at least on some level, that some things are just morally wrong, and should not be done. I'll take an easy example: we have the technology to safely and painlessly sterilize people who have congenital mental or emotional defects. Such people obviously aren't capable of making rational decisions about reproduction by themselves, due to their defects, and we have the technology to do it for them. Should we do it?
The correct answer here is no. No person has the right to do something that drastic to another without just cause and without that person's informed consent. So some things are simply morally wrong. (You don't have to agree, but you do have to have an opinion. Not having an opinion on this question means you have no ethical sense at all; in that case, just stop reading, because I'm not interested in arguing about the nature of ethics with you.)
Is cloning wrong, morally, ethically, pragmatically, or for some other reason? How about stem cell research using in vitro embryos? I don't have answers to those questions, but it's vitally important that we ask them. Because the answer might just turn out to be yes. And if it is, and we didn't bother to think about it before acting, the results would be tragic beyond any justification.
When you were small, your parents-- or somebody, surely-- taught you to look both ways before crossing a road. This is the same principle. Should we ban cloning, or nanotechnology, or any such thing? I don't know. But I know, beyond a shadow of a doubt, that we must ask the questions, and we must have the arguments, because the risk of acting without forethought is far too great.
Also, nice to know there might be millions of natural deadly bacteria in it.
Jellies and jams, and other canned and jarred goods, are inherently pasteurized. The jelly is poured into the jar while still quite hot-- over 140 F-- and the jar sealed. No bacteria in a jelly jar unless the seal is broken. You don't have to be afraid of the jelly jar any more.
Your idea is, in a word, dumb. It is simply not possible to design a nanotechnological weapon that will kill only whom you want it to kill. Can't be done. What are you envisioning, here, some kind of nanobot cloud that goes around peeking at people's driver's licenses to see if they're on the "kill" list or the "don't kill" list?
The weapon that comes closest to perfection is now the same thing that it always has been: an well-trained and well-armed infantryman. Or, even better, a couple million of them.
Re:They're against it because he's for it?
on
Don't Stymie Nanotech
·
· Score: 3, Insightful
He thinks the constitution guarantees the right to overthrow the government through armed struggle.
Oh, not really. He takes the assumption that the right of armed revolution is a given, which is fair considering that the concept is basically codified in the Declaration of Independence. But the bulk of his paper refers to Tennessee state case law, which actually has supported the idea of the right of the citizenry to possess arms for the purpose (among others) of resisting oppression should it arise. Frankly, it's a pretty interesting idea in the age-old gun control argument.
So, why does he oppose even modest regulation?
Short answer: because there must always be those who favor total regulation, and those who favor no regulation at all, so that the rest of us can adopt the measured approach of some regulation.
Compromise can't happen unless people disagree. I salute the Libertarian nutjobs out there, because they're doing us the service of reminding us why some encroachment on freedom is necessary in a free society. And, bless their little hearts, you've just gotta respect people who stick to it even though they never, ever get their way.
The only reason we humans haven't destroyed the world is that we have the brains to restrain ourselves
Huh? The only reason we humans haven't destroyed the world is that we're physically incapable of doing so. This planet is bigger than we are. Don't underestimate it.
We would eventually destroy the earth for the purposes of supporting most animal life if we continued down the path we started in the industrial revolution.
There's no evidence at all to support this statement. It is possible to introduce substances into the soil or the water that prevent plant life from growing, but can you imagine what it would take to cover the entire surface of the planet with such substances? It's simply not possible for humanity, even if we put our minds to it, to manufacture that much of anything. We might mess up a few places here and there, but the vast majority of the biosphere would survive just fine. Ten thousand years or so-- the blink of an eye in geological terms-- and you won't even be able to see the bare spots.
If you need convincing of this, just remember that life-- complex life, not just microbes-- thrives around deep ocean volcanic vents. A less hospitable environment would be hard for us to imagine: superheated, saturated with heavy metals and other unhealthy substances, and yet crab-like arthropods and other complex animal forms survive down there with no trouble at all.
The world is bigger, older, and more complex than you give it credit for being.
I feel I should point out, as I did in another post, that plants do exactly what you're talking about. They consume air, sunlight, water, and dirt (not mineral ore alone, but the good organic stuff that covers the surface of the planet to a depth of several feet) and use them to grow and reproduce.
And they have, essentially, reproduced out of control. Every place on Earth where a plant could grow, there's one growing.
Except that goddamned bare spot in my lawn. But that's another issue altogether.
Re:It doesn't take half a brain to see this.
on
Don't Stymie Nanotech
·
· Score: 3, Interesting
Withstanding any climate is terribly easy when you're made of molecules of metals and/or minerals.
No, ma'am, it's not. Molecules are basically all the same when it comes to interacting with their environment. It's only when you start to look at them on the macroscopic scale that you start to see factors like hardness and tensile strength come into play. At the atomic level, a molecule composed of iron and lead is just as fragile as one made of nitrogen and oxygen. The same outside factors-- temperature, pressure, pH, radiation, etc.-- can crack metallic molecules apart as easily as anything else.
That said, nobody has proposed molecule-scale structures made of metal atoms. Carbon is just too damn useful not to construct the basic structure of your nanotechnological machine out of it. Once you start thinking about these things, the realization dawns that the most suitable elements for molecule-scale machines-- carbon, nitrogen, phosphorous, oxygen-- are the same elements that comprise all life on Earth. Maybe there's a reason for that...
We can't 'eat' air and dirt and use it (and it alone) to grow and reproduce.
Uh... what do you think a plant is? A plant uses nothing but air, dirt, sunlight, and water to grow and reproduce. What's the result? Little bastards are everywhere, man!
Re:It has more benefits than drawbacks...
on
Don't Stymie Nanotech
·
· Score: 3, Insightful
Hell, at first, they weren't even sure if an a-bomb detonation would IGNITE THE ATMOSPHERE, killing us all. Luckily, it didn't-- we dodged a bullet.
That's kind of like saying, "This morning I got out of bed, had my oatmeal, and went to work, all without getting gored by a unicorn! Whew! That was close! Dodged a bullet there."
The idea of self-replicating nanotechnological assemblers is a dumb one, and Drexler deserves a special form of ridicule for ever seriously proposing it. That said, though, the "gray goo" problem is already here, and it's widespread. Except it's not gray. It's green.
Everybody is uninformed about nanotechnology. It's a term like "theology" or "philosophy," too broad to have any real meaning.
That said, let me summarize what I know about nanotechnology so you can decide if I'm insufficiently informed. It all started with Feynman's APS talk back in '59. If I remember correctly, it was entitled, "There's Room at the Bottom," or something like that. In it, he talked about the theoretical basis for molecule-scale structures: manufacturing through evaporation, the challenge of lubrication, and so on. Interestingly, I seem to recall, Feynman essentially ignored the implications of the Uncertainty Principle in his talk. That may be my imagination, though; it's been a long time.
Meanwhile, von Neumann was doing theoretical work of his own on self-replicating systems. (His work actually predated Feynman's talk by several years, but that's close enough to merit a "meanwhile" from me.)
Drexler first put the ideas together in a serious way in 1981, and in greater detail in his seminal '86 book, Engines of Creation. (I sold my copy years ago to a used book store, so don't expect any chapter-and-verse quotes from me.) He postulated self-replicating devices for manipulating atoms individually; he called them "assemblers." If I remember correctly, he also coined the term "gray goo" to refer to the nightmare of a runaway assembler that devours all available raw materials to manufacture more copies of itself, burying the surface of the Earth in a homogenous sludge.
Since the 80's, Drexler and others have done a mountain of work on nanotechnological ideas, most of them centering around the idea of the atomic-scale self-reproducing assembler. But that's not the whole story.
Back to the 1970's. We have these two basic ideas: atomic-scale manufacturing (Feynman) and self-replicating machines (von Neumann). Drexler jumped to the conclusion that these two ideas can be made to work together and ran with it. But the blanket term "nanotechnology" has since been applied to any non-biological physical process that occurs on the nanometer scale, not just Drexler's blue-sky ideas. That's why I say the term "nanotechnology" is essentially useless in any sort of technical discussion. Electron microscopy is nanotechnology. The synthesis of drugs is nanotechnology. PCR amplification of DNA is nanotechnology. Electroplating is nanotechnology. Drexler's self-replicating assemblers are nanotechnology. Everything is nanotechnology, in one way or another. And some ideas that can fairly be called nanotechnology are... well, let's just say they're a hell of a lot less plausible than others.
To take a specific example, Drexler's ideas of atomic-scale assemblers that replicate themselves and also assemble other atomic-scale structures are here already. They're called enzymes, and they're everywhere. The problem is that they only work inside a narrow range of temperature and environment. If the pH is too high or too low, the enzymes-- or "assemblers," if you prefer-- simply don't work. So they have to be contained inside little self-regulating environment bubbles: cells. And cells-- well, most kinds, anyway-- are too fragile to exist for long without external support. Thus, organisms. And even when an enzyme is in the perfect environment, contained inside a cell that's in turn protected by an entire organism, it still only works about half the time. Even something as seemingly harmless as sunlight can attack enzymes like artillery shells, blasting those fragile molecules into pieces before they can do their jobs. But the biological processes are so ridiculously redundant that 50% or more is an acceptable rate of failure.
Drexler envisions a very clean, precise atomic-scale manufacturing process. He assumes that this is possible because we're talking about putting atoms in place one at a time; there's no reason any nanotechnologically manufactured object should ever have a flaw in it. But, while that's theoretically possible, it's a lot harder to achieve than you might think. Remember what I said about temperature and environment? Nanotechnological-- or, if you prefer, biological, for at this point they're the same thing-- processes are fragile and delicate.
So jumping to the conclusion that we will have nanotechnology is meaningless, because nanotechnology means any number of things that cover the spectrum from the mundane to the fanciful.
The possible flaw is that by the time they get the technology necessary to live in space sustainably long-term, mature nanotechnology will be available.
You appear to have confused science fiction with reality. There's no context in which a statement like "nanotechnology will be available" (emphasis mine) can be taken seriously. Apart from the fact that the word "nanotechnology," by itself, is too broad to have any relevance... oh, wait.
And once the singularity happens all bets are off, but chances are nanobot probes will be heading off in all directions at close to the speed of light, which means their ship will more than likely get infected, unless this singularity is benign.
Sorry, I should have read your whole post before responding. I didn't realize until after I'd already hit "reply" that you're a loony.
I found this little gem in the "Oceania Project" farewell letter.
In retrospect, the biggest problem concerning The Atlantis Project was lack of interest. Lack of interest and the fact that its precepts were based in Libertarian politics.
The two problems concerning the Project were lack of interest and the fact that its precepts were based in Libertarian politics. And a lack of funding.
The three problems concerning the Project were lack of interest, the fact that its precepts were based in Libertarian politics, and a lack of funding. And an almost fanatical devotion to our cause.
The four... no. Amongst the problems... amongst the challenges... are such factors as a general lack of interest....
I'll come in again.
I'm not sure I'm ready to put the responsibility for saving mankind in the hands of a guy who's seen too many Monty Python sketches.
(No sense of humor? It's a joke. I'm kidding. The first two sentences are actually in the letter. I added the rest, because it's funny. Ha-ha.)
I've ranted about this before. De jure standards have marginal relevance at best. If your web page is designed for public consumption, like a company or personal web site or some such, the correct thing to do is design it so that it renders correctly in IE. If that page fails to render correctly in Mozilla, so what? With IE holding something like 95% of the browser market, the odds that a random visitor to your site is going to be running IE 5 or 6 are overwhelming. It makes little sense, then, to design your site to conform to de jure standards while rendering incorrectly for 95% of all visitors.
This is, ultimately, why Mozilla is irrelevant. It's bloated, slow, enormous to download, amateurish, and ugly, but it would be possible to overcome all of those thing given sufficient time and effort. The problem, the last nail in the coffin if you will, is that Mozilla fails to render many web sites that IE renders perfectly. I'm sure the problem lies with the Mozilla DOM implementation, but the details aren't important. What's important is that it just doesn't work.
When Mozilla is as fast as IE for Windows, when it has a native UI, and most importantly when it has 100% compatibility with IE 5 and 6, then and only then will it become a reasonable alternative. Until then, it'll never achieve any sort of widespread adoption because it will have nothing at all to offer over the de facto standard.
Microsoft has finished product? I've never seen a piece of microsoft "finished product" that wasn't released at sub beta quality.
I suppose you can do better, then. Please provide us with information-- links will do, we're not picky-- describing any software project that you've completed that rivals the functionality or quality of, say, Microsoft Office XP.
IE won because it came with the OS. It's that simple.
No, it isn't. Back before IE 4.0, Netscape was a far more popular browser, despite the fact that IE came bundled with whichever Windows version was shipping at that time. (I forget.)
As IE got better and better, Netscape got worse and worse, and then stopped altogether. Whereas people had been downloading Netscape in ever-increasing numbers, they simply stopped when IE became a better alternative.
IE won because Microsoft kept making it better, while Netscape kept making their browser worse.
None of the rest of the stuff in your comment means anything because you prefaced it with, "I feel I have to badmouth them at every opportunity just to even the scales." I appreciate you throwing in that little remark, so I wouldn't waste any time reading the rest of your comment.
Because it's integrated into the OS, meaning that parts of it have already started before you run it. Comparing its speed to the speed of a browser that isn't integrated is comparing apples with oranges.
Sounds like you're confused. You're talking about startup speed, as in the time required to open a new instance. We're talking about actual running speed, as in the time required to render a given web page. IE is faster than other browsers.
Given the state of Mozilla/Phoenix nowadays, that's debatable.
No, it isn't. Mozilla is still as slow, bloated, and bug-ridden as it was in the pre-1.0 days. In fact, I've just about come to the conclusion that the 1.0 release was just a big joke. They took a random nightly build and slapped the 1.0 label on it and sent out a press release.
There's no reason in the world Grandma couldn't use Mozilla on Linux for e-mail and web surfing.
Assuming you're ready to do lots and lots of stuff for Grandma, sure. But if you'd rather not spend your free time tinkering with her computer, or if she lives far away from you, I'd think twice. Unless you enjoy talking your grandmother through "rpm -U" over the phone.
You can buy this machine and run redhat out of the box.
Are you absolutely positive about this? Like, have independent confirmation on this from a knowledgeable SGI employee? Because I'm almost completely certain that this is not, in fact, even remotely true. My understanding, which is only slightly out of date, is that you will have to run a kernel built with SGI's patches, of which there are many.
But if my info is wrong, I'd like to be corrected by somebody who knows better than I do.
There are basically three ways to parallelize an application: processes, threads, and sprocs or other lightweight shared-memory process implementations.
Processes for multiprocessing suck. The reasons for this are well known and won't be re-stated here, but cross-reference "shared memory."
Threads are standardized, light-weight, and efficient.
Sprocs are an IRIX-specific thing, but there are similar dohickeys on other OS platforms. A sproc is basically another instance of a process that shares its address space. Sprocs are similar to threads in a lot of ways, but they're not as elegant for many applications and they're highly platform-specific. Programming with this is tough, and porting virtually impossible.
Threads-- and, almost as important, thread performance-- are critically important for application scalability under any operating system, UNIX or otherwise.
WTF are you talking about? SGI does not own the original UNIX trademark. The Open Group owns that trademark. SGI has gotten various versions of IRIX certified by The Open Group as conforming to various UNIX standards, and that's why they're entitled to call IRIX "UNIX."
Slashdot Mirror
1) Put a cap on how many copies of itself a bot can make.
This does nothing to prevent a "gray goo" problem. All it takes is one "mutation"-- an error in manufacturing-- to remove this cap. In biological organisms, this is called "cancer," and it's something of a problem.
2) Add some sort of mechanism to the bots to only allow duplication when there are less than some number of bots in the surrounding area.
So rather than a hard limit on reproductions, you have an environmental feedback mechanism. Again, one mutation disables this feature, or worse, inverts it. Now the assembler only reproduces when the concentration in the local environment is greater than a given value.
3) Only allow bots to reproduce when told to do so by Humans.
Great idea. At this point, they're not self-assembling any more. This is, in fact, the only solution to the problem.
Of course, if the assemblers are able to self-assemble, it's quite likely that they will be caused to self-assemble by some external factor sooner or later. So the only safe way to apply this solution is to make the assemblers physically incapable of assembling themselves.
4) provide som sort of "master kill signal" (i.e. a specific radio frequency) that will cause all bots reciving it to cease functioning.
Just as it only takes one mutation-- or manufacturing error-- to disable a reproductive limit, so too does it take only one error to disable this idea. Too dangerous.
5) Any of several other ideas I didn't just think of in the past 2 min.
I hope the ones you think of in the next 2 minutes are somewhat better that these ones. For inspiration, look at living organisms, and observe all the things that can go wrong with them. Don't suggest anything that leads to one of those states.
And how do we know the goo would be grey, anyway?
The idea of a "gray goo" is favored by people who don't realize that nanotechnology is an old, old idea. There's already goo all over the place, but it isn't gray. It's green.
I'm stunned that four people felt compelled to give you mod points. Your comment is the most absurd one I've seen so far in this thread, and that's counting the trolls.
The wonderful thing about being a human being is that we can choose what to do and what not to do. I can choose to stop at a crosswalk, or I can choose to plow through a crowd of second-graders.
Your argument is basically that it's foolish to stop. Somebody out there is going to plow through a crosswalk anyway, so we might as well get in there and figure out how to use it to our advantage. If you think I'm misrepresenting your argument, maybe you'd better go back and read your own words again. "New technologies will come and be used whether we like it or not. Cloning will occur whether we ban it or not." You say this as if there were no moral or ethical aspect to it whatsoever, and that's simply not true.
Some things just should not be done. If you're an amoral person-- and your post, especially the part about "religious paranoid idiots," certainly seems to suggest that you are-- then you probably reject this assertion on its face. If that's the case, I won't bother trying to convince you otherwise. (My opinion is that people with no sense of morality or ethics at all are mentally ill in a way we just haven't figured out yet. Nothing personal; it's just a theory.)
So let's just take as given that you believe, at least on some level, that some things are just morally wrong, and should not be done. I'll take an easy example: we have the technology to safely and painlessly sterilize people who have congenital mental or emotional defects. Such people obviously aren't capable of making rational decisions about reproduction by themselves, due to their defects, and we have the technology to do it for them. Should we do it?
The correct answer here is no. No person has the right to do something that drastic to another without just cause and without that person's informed consent. So some things are simply morally wrong. (You don't have to agree, but you do have to have an opinion. Not having an opinion on this question means you have no ethical sense at all; in that case, just stop reading, because I'm not interested in arguing about the nature of ethics with you.)
Is cloning wrong, morally, ethically, pragmatically, or for some other reason? How about stem cell research using in vitro embryos? I don't have answers to those questions, but it's vitally important that we ask them. Because the answer might just turn out to be yes. And if it is, and we didn't bother to think about it before acting, the results would be tragic beyond any justification.
When you were small, your parents-- or somebody, surely-- taught you to look both ways before crossing a road. This is the same principle. Should we ban cloning, or nanotechnology, or any such thing? I don't know. But I know, beyond a shadow of a doubt, that we must ask the questions, and we must have the arguments, because the risk of acting without forethought is far too great.
Also, nice to know there might be millions of natural deadly bacteria in it.
Jellies and jams, and other canned and jarred goods, are inherently pasteurized. The jelly is poured into the jar while still quite hot-- over 140 F-- and the jar sealed. No bacteria in a jelly jar unless the seal is broken. You don't have to be afraid of the jelly jar any more.
Your idea is, in a word, dumb. It is simply not possible to design a nanotechnological weapon that will kill only whom you want it to kill. Can't be done. What are you envisioning, here, some kind of nanobot cloud that goes around peeking at people's driver's licenses to see if they're on the "kill" list or the "don't kill" list?
The weapon that comes closest to perfection is now the same thing that it always has been: an well-trained and well-armed infantryman. Or, even better, a couple million of them.
He thinks the constitution guarantees the right to overthrow the government through armed struggle.
Oh, not really. He takes the assumption that the right of armed revolution is a given, which is fair considering that the concept is basically codified in the Declaration of Independence. But the bulk of his paper refers to Tennessee state case law, which actually has supported the idea of the right of the citizenry to possess arms for the purpose (among others) of resisting oppression should it arise. Frankly, it's a pretty interesting idea in the age-old gun control argument.
So, why does he oppose even modest regulation?
Short answer: because there must always be those who favor total regulation, and those who favor no regulation at all, so that the rest of us can adopt the measured approach of some regulation.
Compromise can't happen unless people disagree. I salute the Libertarian nutjobs out there, because they're doing us the service of reminding us why some encroachment on freedom is necessary in a free society. And, bless their little hearts, you've just gotta respect people who stick to it even though they never, ever get their way.
The only reason we humans haven't destroyed the world is that we have the brains to restrain ourselves
Huh? The only reason we humans haven't destroyed the world is that we're physically incapable of doing so. This planet is bigger than we are. Don't underestimate it.
We would eventually destroy the earth for the purposes of supporting most animal life if we continued down the path we started in the industrial revolution.
There's no evidence at all to support this statement. It is possible to introduce substances into the soil or the water that prevent plant life from growing, but can you imagine what it would take to cover the entire surface of the planet with such substances? It's simply not possible for humanity, even if we put our minds to it, to manufacture that much of anything. We might mess up a few places here and there, but the vast majority of the biosphere would survive just fine. Ten thousand years or so-- the blink of an eye in geological terms-- and you won't even be able to see the bare spots.
If you need convincing of this, just remember that life-- complex life, not just microbes-- thrives around deep ocean volcanic vents. A less hospitable environment would be hard for us to imagine: superheated, saturated with heavy metals and other unhealthy substances, and yet crab-like arthropods and other complex animal forms survive down there with no trouble at all.
The world is bigger, older, and more complex than you give it credit for being.
I feel I should point out, as I did in another post, that plants do exactly what you're talking about. They consume air, sunlight, water, and dirt (not mineral ore alone, but the good organic stuff that covers the surface of the planet to a depth of several feet) and use them to grow and reproduce.
And they have, essentially, reproduced out of control. Every place on Earth where a plant could grow, there's one growing.
Except that goddamned bare spot in my lawn. But that's another issue altogether.
Withstanding any climate is terribly easy when you're made of molecules of metals and/or minerals.
No, ma'am, it's not. Molecules are basically all the same when it comes to interacting with their environment. It's only when you start to look at them on the macroscopic scale that you start to see factors like hardness and tensile strength come into play. At the atomic level, a molecule composed of iron and lead is just as fragile as one made of nitrogen and oxygen. The same outside factors-- temperature, pressure, pH, radiation, etc.-- can crack metallic molecules apart as easily as anything else.
That said, nobody has proposed molecule-scale structures made of metal atoms. Carbon is just too damn useful not to construct the basic structure of your nanotechnological machine out of it. Once you start thinking about these things, the realization dawns that the most suitable elements for molecule-scale machines-- carbon, nitrogen, phosphorous, oxygen-- are the same elements that comprise all life on Earth. Maybe there's a reason for that...
We can't 'eat' air and dirt and use it (and it alone) to grow and reproduce.
Uh... what do you think a plant is? A plant uses nothing but air, dirt, sunlight, and water to grow and reproduce. What's the result? Little bastards are everywhere, man!
Hell, at first, they weren't even sure if an a-bomb detonation would IGNITE THE ATMOSPHERE, killing us all. Luckily, it didn't-- we dodged a bullet.
That's kind of like saying, "This morning I got out of bed, had my oatmeal, and went to work, all without getting gored by a unicorn! Whew! That was close! Dodged a bullet there."
The idea of self-replicating nanotechnological assemblers is a dumb one, and Drexler deserves a special form of ridicule for ever seriously proposing it. That said, though, the "gray goo" problem is already here, and it's widespread. Except it's not gray. It's green.
You obviously are uninformed about nanotechnolgy.
Everybody is uninformed about nanotechnology. It's a term like "theology" or "philosophy," too broad to have any real meaning.
That said, let me summarize what I know about nanotechnology so you can decide if I'm insufficiently informed. It all started with Feynman's APS talk back in '59. If I remember correctly, it was entitled, "There's Room at the Bottom," or something like that. In it, he talked about the theoretical basis for molecule-scale structures: manufacturing through evaporation, the challenge of lubrication, and so on. Interestingly, I seem to recall, Feynman essentially ignored the implications of the Uncertainty Principle in his talk. That may be my imagination, though; it's been a long time.
Meanwhile, von Neumann was doing theoretical work of his own on self-replicating systems. (His work actually predated Feynman's talk by several years, but that's close enough to merit a "meanwhile" from me.)
Drexler first put the ideas together in a serious way in 1981, and in greater detail in his seminal '86 book, Engines of Creation. (I sold my copy years ago to a used book store, so don't expect any chapter-and-verse quotes from me.) He postulated self-replicating devices for manipulating atoms individually; he called them "assemblers." If I remember correctly, he also coined the term "gray goo" to refer to the nightmare of a runaway assembler that devours all available raw materials to manufacture more copies of itself, burying the surface of the Earth in a homogenous sludge.
Since the 80's, Drexler and others have done a mountain of work on nanotechnological ideas, most of them centering around the idea of the atomic-scale self-reproducing assembler. But that's not the whole story.
Back to the 1970's. We have these two basic ideas: atomic-scale manufacturing (Feynman) and self-replicating machines (von Neumann). Drexler jumped to the conclusion that these two ideas can be made to work together and ran with it. But the blanket term "nanotechnology" has since been applied to any non-biological physical process that occurs on the nanometer scale, not just Drexler's blue-sky ideas. That's why I say the term "nanotechnology" is essentially useless in any sort of technical discussion. Electron microscopy is nanotechnology. The synthesis of drugs is nanotechnology. PCR amplification of DNA is nanotechnology. Electroplating is nanotechnology. Drexler's self-replicating assemblers are nanotechnology. Everything is nanotechnology, in one way or another. And some ideas that can fairly be called nanotechnology are... well, let's just say they're a hell of a lot less plausible than others.
To take a specific example, Drexler's ideas of atomic-scale assemblers that replicate themselves and also assemble other atomic-scale structures are here already. They're called enzymes, and they're everywhere. The problem is that they only work inside a narrow range of temperature and environment. If the pH is too high or too low, the enzymes-- or "assemblers," if you prefer-- simply don't work. So they have to be contained inside little self-regulating environment bubbles: cells. And cells-- well, most kinds, anyway-- are too fragile to exist for long without external support. Thus, organisms. And even when an enzyme is in the perfect environment, contained inside a cell that's in turn protected by an entire organism, it still only works about half the time. Even something as seemingly harmless as sunlight can attack enzymes like artillery shells, blasting those fragile molecules into pieces before they can do their jobs. But the biological processes are so ridiculously redundant that 50% or more is an acceptable rate of failure.
Drexler envisions a very clean, precise atomic-scale manufacturing process. He assumes that this is possible because we're talking about putting atoms in place one at a time; there's no reason any nanotechnologically manufactured object should ever have a flaw in it. But, while that's theoretically possible, it's a lot harder to achieve than you might think. Remember what I said about temperature and environment? Nanotechnological-- or, if you prefer, biological, for at this point they're the same thing-- processes are fragile and delicate.
So jumping to the conclusion that we will have nanotechnology is meaningless, because nanotechnology means any number of things that cover the spectrum from the mundane to the fanciful.
Just because the cigarette smoke, BO and other such cruft makes it absolutely unbarable to breath doesn't make it a vacume.
Is the "e" key on your keyboard broken? It seems to be failing when you need it, and then firing off at random when you don't.
Might want to have that looked at.
Whats the secret trick to get my laptop above 1024x768?
It's easy. Just click here and the press the green button.
The possible flaw is that by the time they get the technology necessary to live in space sustainably long-term, mature nanotechnology will be available.
You appear to have confused science fiction with reality. There's no context in which a statement like "nanotechnology will be available" (emphasis mine) can be taken seriously. Apart from the fact that the word "nanotechnology," by itself, is too broad to have any relevance... oh, wait.
And once the singularity happens all bets are off, but chances are nanobot probes will be heading off in all directions at close to the speed of light, which means their ship will more than likely get infected, unless this singularity is benign.
Sorry, I should have read your whole post before responding. I didn't realize until after I'd already hit "reply" that you're a loony.
Carry on.
(No sense of humor? It's a joke. I'm kidding. The first two sentences are actually in the letter. I added the rest, because it's funny. Ha-ha.)
I've ranted about this before. De jure standards have marginal relevance at best. If your web page is designed for public consumption, like a company or personal web site or some such, the correct thing to do is design it so that it renders correctly in IE. If that page fails to render correctly in Mozilla, so what? With IE holding something like 95% of the browser market, the odds that a random visitor to your site is going to be running IE 5 or 6 are overwhelming. It makes little sense, then, to design your site to conform to de jure standards while rendering incorrectly for 95% of all visitors.
This is, ultimately, why Mozilla is irrelevant. It's bloated, slow, enormous to download, amateurish, and ugly, but it would be possible to overcome all of those thing given sufficient time and effort. The problem, the last nail in the coffin if you will, is that Mozilla fails to render many web sites that IE renders perfectly. I'm sure the problem lies with the Mozilla DOM implementation, but the details aren't important. What's important is that it just doesn't work.
When Mozilla is as fast as IE for Windows, when it has a native UI, and most importantly when it has 100% compatibility with IE 5 and 6, then and only then will it become a reasonable alternative. Until then, it'll never achieve any sort of widespread adoption because it will have nothing at all to offer over the de facto standard.
Yup. Every single one of those examples is a finished product. Well, except the "completely new Win" thing that you mentioned.
Microsoft has finished product? I've never seen a piece of microsoft "finished product" that wasn't released at sub beta quality.
I suppose you can do better, then. Please provide us with information-- links will do, we're not picky-- describing any software project that you've completed that rivals the functionality or quality of, say, Microsoft Office XP.
Ready... go.
IE won because it came with the OS. It's that simple.
No, it isn't. Back before IE 4.0, Netscape was a far more popular browser, despite the fact that IE came bundled with whichever Windows version was shipping at that time. (I forget.)
As IE got better and better, Netscape got worse and worse, and then stopped altogether. Whereas people had been downloading Netscape in ever-increasing numbers, they simply stopped when IE became a better alternative.
IE won because Microsoft kept making it better, while Netscape kept making their browser worse.
None of the rest of the stuff in your comment means anything because you prefaced it with, "I feel I have to badmouth them at every opportunity just to even the scales." I appreciate you throwing in that little remark, so I wouldn't waste any time reading the rest of your comment.
Because it's integrated into the OS, meaning that parts of it have already started before you run it. Comparing its speed to the speed of a browser that isn't integrated is comparing apples with oranges.
Sounds like you're confused. You're talking about startup speed, as in the time required to open a new instance. We're talking about actual running speed, as in the time required to render a given web page. IE is faster than other browsers.
Given the state of Mozilla/Phoenix nowadays, that's debatable.
No, it isn't. Mozilla is still as slow, bloated, and bug-ridden as it was in the pre-1.0 days. In fact, I've just about come to the conclusion that the 1.0 release was just a big joke. They took a random nightly build and slapped the 1.0 label on it and sent out a press release.
There's no reason in the world Grandma couldn't use Mozilla on Linux for e-mail and web surfing.
Assuming you're ready to do lots and lots of stuff for Grandma, sure. But if you'd rather not spend your free time tinkering with her computer, or if she lives far away from you, I'd think twice. Unless you enjoy talking your grandmother through "rpm -U" over the phone.
No idea. You can learn more about sprocs by starting with the man page, here.
Maybe someone can describe the bricks?
Yeah, maybe.
You can buy this machine and run redhat out of the box.
Are you absolutely positive about this? Like, have independent confirmation on this from a knowledgeable SGI employee? Because I'm almost completely certain that this is not, in fact, even remotely true. My understanding, which is only slightly out of date, is that you will have to run a kernel built with SGI's patches, of which there are many.
But if my info is wrong, I'd like to be corrected by somebody who knows better than I do.
You seem to be completely out of your mind.
There are basically three ways to parallelize an application: processes, threads, and sprocs or other lightweight shared-memory process implementations.
Processes for multiprocessing suck. The reasons for this are well known and won't be re-stated here, but cross-reference "shared memory."
Threads are standardized, light-weight, and efficient.
Sprocs are an IRIX-specific thing, but there are similar dohickeys on other OS platforms. A sproc is basically another instance of a process that shares its address space. Sprocs are similar to threads in a lot of ways, but they're not as elegant for many applications and they're highly platform-specific. Programming with this is tough, and porting virtually impossible.
Threads-- and, almost as important, thread performance-- are critically important for application scalability under any operating system, UNIX or otherwise.
WTF are you talking about? SGI does not own the original UNIX trademark. The Open Group owns that trademark. SGI has gotten various versions of IRIX certified by The Open Group as conforming to various UNIX standards, and that's why they're entitled to call IRIX "UNIX."