You might be able to make sub-atomic switching devices, but they wouldn't properly be called "transistors." Then again, these might be stretching the term a bit. At least they work on electrons, though.
Mechanical computers would be smaller then anything you could design to process electrons.
I remember, it was probably close to a decade ago, that someone had synthesized a small X-shaped molecule that should function as a diode. IIRC, it was less than 20 atoms. Plenty of people are talking about molecular diodes.
There is no way that you're going to move around some 20-atom monster at the same rate you can move around an electron, or even a thousand electrons. It's a simple matter of mass ratios: electrons are light, molecules are heavy.
If you actually read Nanosystems, you'll see that he's talking about ~1 GHz systems. We've already surpassed that with microelectronics. Nanoelectronics will be probably be smaller, simpler, and more efficient than nanomechanical computer systems and thousands if not millions of times faster. Of course, nano-optical systems, or something weird we can't guess at yet, might be even faster.
Yes, your vague half-truths are very valuable.
on
Babbage, A Look Back
·
· Score: 1
The written zero came through the middle-east, not from it. It's a matter of historical record that it was used for over a hundred years in India before being brought to the middle-east; hundreds of years before that it was independently invented by the Mayans, but they were isolated and it didn't spread from them. At any rate, the concept of zero, and modern digital numbers, have been around much longer, embodied in the abacus; there's no saying how long, as the earliest abacii were no doubt just organized piles of ordinary stones. Modern digits, using zero, aren't what I meant, though.
You've been running around with a misconception, and apparently looking down on others who don't hold it. Now that it's corrected, are you a better programmer? I don't believe it's done me any good, it's just a bit of trivia I picked up along the way.
I'm not against history, I'm against educational snobbery: a form of ignorance that causes you to measure the effectiveness of a person at tasks you are good at by the similarity of your education, regardless of which parts of your education were actually relevant to the tasks in question.
A hundred years ago, you'd be ridiculed in similar fashion for not knowing Latin. Now nobody considers it part of basic education, just an obscure subject for specialists. A decline in human knowledge? Nonsense! There are more scholars in more obscure fields now than ever before. By focusing on what is actually relevant to a field, we have more time and energy to make progress in it, and more freedom to follow our own tastes so a team of people will have a broader background of knowledge to draw on, instead of everyone carrying the same limited education around with them.
How does the saying go? "Not constant addition, constant subtraction: strip away the inessentials to approach perfection."
I really don't think Drexler is seriously proposing such a design. He just used the mechanical computer in Nanosystems as an easy example to demonstrate feasibility. Electronic systems would be much faster and smaller, it's just that the physics is much harder to explain convincingly without having models to fiddle with.
The history of our trade is a glorious thing, full of great men and brilliant engineering. Only through it's study can we hope to go as far as they did.
Absolute garbage. They didn't advance mathematics and computer science by studying history, and we won't either. To continue their work it's important to learn the things they learned, not to learn about them as people.
Where would it stop? Why is someone an ignorant fool for not knowing or caring about this or that person in particular? Which historical figures are essential?
An answer I'm hearing a lot is the snob's answer: "I know what I should, people who know less are ignorant (people who know more are indulging themselves in trivia)."
The real place necessity ends is at the start: it's all trivia. Just as you don't need to know Leibniz or Newton to understand calculus, you don't need to know Grace Hopper to make a compiler. Science history is an interesting subject, but it's no prerequisite for science.
Does it hurt you to be unaware of who first thought of writing numbers as digits? The genius who came up with reliable procedures for adding them? The spectacular genius who generalized expression of fractions by numbers?
No, they've been lost in the mists of time. Most people aren't bothered by this, as it's inconceivable in all but the most abstract sense that there was a time when no human knew these things. Sometimes I think that's why they're so commonly learned: they are no longer the territory of A Great Man and his disciples, just common knowledge a hair above common sense. Certainly, nobody who tackles these subjects only to be confronted by obnoxious fools with, "How dare you learn only the timeless mathematical truth and ignore the circumstances of its discovery?!"
In one place it says: "The most important things the shuttle group does -- carefully planning the software in advance, writing no code until the design is complete, making no changes without supporting blueprints, keeping a completely accurate record of the code -- are not expensive."
And in another: "on a dollars-per-line basis, it [the group is] among the nation's most expensive software organizations."
Of course, those things are exactly what make it so expensive. It doesn't sound expensive because it's just work. Only, it's a vast amount of work, done by highly-paid professionals. Another bit mentions a change to 6366 lines of code, that was specified by 2500 pages of documentation.
The subject of their efforts is 420,000 lines of code. They've had over 20 years, with a budget currently at 32 million dollars per year. That's enough money for man-weeks for each line of code!
Fixed prices, subscription quotas, and the threat of withholding (effectively destroying) finished work?
It's complex, inflexible, and requires too much administration and too large a commitment. Can you see yourself signing up for donation subscriptions to support a hundred software projects? Surely you benefit from the efforts of at least that many.
I believe that the best solution is the simplest: just give money to people who have made the stuff you like, don't make them screw around with their own weird variant to make it sound like something other than donation. Don't withhold money from one that is already profitable by the donations of others; huge profits mean that others will be attracted to compete in the area, that's how capitalism allocates resources efficiently over the long term.
If you do that, then people can just get down to the work of figuring out what you want, and making it, knowing from past experience that you can be trusted to make such efforts profitable without coersion.
I left out the part about how, because due to time dilation, he will not be truly dead for several million years at least, the usual inheritance tax was not levied on his estate.
Stalin, while relaxing in a tropical amusement park, picking his nose, was struck by a coconut, prompting a brilliant idea! He immediately dragged his car (he couldn't get any wheel's, because they were patented) to his workshop and made a set of airproof underpants. He considered using charcoal filter's to absorb fart's, but because it would have to be changed, he instead leveraged his recent singularity research to create a small, contained black hole.
To test them, knowing that adult's might be too polite to be honest or too dignified to participate, he gathered together a group of children. The test's were held in a shower to handle any accidents that might result from forced farting. It worked brilliantly. The children tried them in turn's, and not an unpleasant whiff escaped.
For a final test, Stalin tried them on himself and stepped into the shower and strained with all his strength. Unfortunately, his mighty blast destabilized the black hole, causing him and the shower curtain to be sucked into it. At first thinking it all part of the fun, the children were overjoyed at the spectacle.
He was awarded an Ig Nobel Prize posthumously, though this is likely no consolation as he is presumed to live in an eternal hell of contained fart's.
(I swear it sounded like a good idea when I started writing...)
Obviously you can't have young hydrogen-only stars unless the universe was different than it is today.
There are other possibilities.
The universe could be infinite and expanding (everything moving farther apart from each other), with new matter appearing in the voids created. If it expands in the right way, effects of a local event can propagate infinitely into new space (populated with new matter) without ever affecting anything that was a beyond a certain distance from the event at the time it happened.
Or the universe could be finite and self-refreshing on large or small scales. Setting aside the big bang/big crunch idea for the moment, matter and energy could "evaporate" at a rate relative to the gravitational constant in the area, to "condense" in areas with relatively low gravity, in a counter-entropic process that produces fresh hydrogen from worn out iron, neutronium, and um... blackholium?
We don't know the nature of our universe, or its origin, and never will be certain that we do. I don't consider natural history to be real science, any more than human history is. They are both certainly worthy pursuits, but are not built on the truly falsifiable claims which characterize science. I mean, you can assume that the laws of physics that work today applied before any human lived, but you can't test it.
I guess my point is that natural history has to be consistent with both current observation and physics, and we know our physics is incomplete. It's a bit early in the game to go chiseling events that happened billions of years ago onto stone tablets and ridiculing dissenters.
They're using rubbing alcohol and liquid oxygen. Despite the fact that low-concentration hydrogen peroxide might be on your bathroom shelf beside the isopropyl alcohol, they are entirely different chemicals.
They'll never get out the exploits like vaccines and cures if they try to develop it in a closed development group. There are many skilled reverse-engineers in biology. Security through obscurity never works.
The thing is, there are limits to copyright. The shorter a work, the harder to defend its copyright in court. For instance, it is impossible to copyright a word, phrase, note, or chord. Short poems, like haiku, push the lower bounds, and have quite weak protection: only a very blatant direct copy might infringe on them.
Obviously, these are not legal (or at least not legally relevant) copyrights, and couldn't be enforced.
I know it's all in fun, but I think it would be more satisfying to mock the system using things that would stand up in court.
Experian is a company that sells crack by spamming millions of schoolchildren. MAPS (the Multidisciplinary Association for Psychedelic Studies) has refused to provide their drug-friendly mailing lists because they insist that crack is not a psychedelic drug. However, Experian threatened to take them to court with the argument that if they can include marijuana under their umbrella, then the definition is broad enough to include crack.
Wisely recognizing that both sides are better off not attracting the attention of the courts, MAPS has apparently backed down.
A loss indeed. You can expect many of your peaceful local potheads to become violent criminal crackheads any day now.
I hope this clears up any misconceptions you may have had from the shamefully vague top-level story. I'm a little fuzzy on some of the details myself, but as usual, trying a few likely domain names gave me access to the essentials.
You might be interested in EROS - the Extremely Reliable Operating System, which takes permissions resolution to its logical extreme: the capability system. If something only needs access to one directory and one port, that's all you give it.
Slashdot Mirror
Now this is a truly beautiful gag. You should work writing technobabble for Star Trek.
But why isn't it moderated "funny"?
You might be able to make sub-atomic switching devices, but they wouldn't properly be called "transistors." Then again, these might be stretching the term a bit. At least they work on electrons, though.
Mechanical computers would be smaller then anything you could design to process electrons.
I remember, it was probably close to a decade ago, that someone had synthesized a small X-shaped molecule that should function as a diode. IIRC, it was less than 20 atoms. Plenty of people are talking about molecular diodes.
There is no way that you're going to move around some 20-atom monster at the same rate you can move around an electron, or even a thousand electrons. It's a simple matter of mass ratios: electrons are light, molecules are heavy.
If you actually read Nanosystems, you'll see that he's talking about ~1 GHz systems. We've already surpassed that with microelectronics. Nanoelectronics will be probably be smaller, simpler, and more efficient than nanomechanical computer systems and thousands if not millions of times faster. Of course, nano-optical systems, or something weird we can't guess at yet, might be even faster.
The written zero came through the middle-east, not from it. It's a matter of historical record that it was used for over a hundred years in India before being brought to the middle-east; hundreds of years before that it was independently invented by the Mayans, but they were isolated and it didn't spread from them. At any rate, the concept of zero, and modern digital numbers, have been around much longer, embodied in the abacus; there's no saying how long, as the earliest abacii were no doubt just organized piles of ordinary stones. Modern digits, using zero, aren't what I meant, though.
You've been running around with a misconception, and apparently looking down on others who don't hold it. Now that it's corrected, are you a better programmer? I don't believe it's done me any good, it's just a bit of trivia I picked up along the way.
I'm not against history, I'm against educational snobbery: a form of ignorance that causes you to measure the effectiveness of a person at tasks you are good at by the similarity of your education, regardless of which parts of your education were actually relevant to the tasks in question.
A hundred years ago, you'd be ridiculed in similar fashion for not knowing Latin. Now nobody considers it part of basic education, just an obscure subject for specialists. A decline in human knowledge? Nonsense! There are more scholars in more obscure fields now than ever before. By focusing on what is actually relevant to a field, we have more time and energy to make progress in it, and more freedom to follow our own tastes so a team of people will have a broader background of knowledge to draw on, instead of everyone carrying the same limited education around with them.
How does the saying go? "Not constant addition, constant subtraction: strip away the inessentials to approach perfection."
I really don't think Drexler is seriously proposing such a design. He just used the mechanical computer in Nanosystems as an easy example to demonstrate feasibility. Electronic systems would be much faster and smaller, it's just that the physics is much harder to explain convincingly without having models to fiddle with.
The history of our trade is a glorious thing, full of great men and brilliant engineering. Only through it's study can we hope to go as far as they did.
Absolute garbage. They didn't advance mathematics and computer science by studying history, and we won't either. To continue their work it's important to learn the things they learned, not to learn about them as people.
Where would it stop? Why is someone an ignorant fool for not knowing or caring about this or that person in particular? Which historical figures are essential?
An answer I'm hearing a lot is the snob's answer: "I know what I should, people who know less are ignorant (people who know more are indulging themselves in trivia)."
The real place necessity ends is at the start: it's all trivia. Just as you don't need to know Leibniz or Newton to understand calculus, you don't need to know Grace Hopper to make a compiler. Science history is an interesting subject, but it's no prerequisite for science.
Does it hurt you to be unaware of who first thought of writing numbers as digits? The genius who came up with reliable procedures for adding them? The spectacular genius who generalized expression of fractions by numbers?
No, they've been lost in the mists of time. Most people aren't bothered by this, as it's inconceivable in all but the most abstract sense that there was a time when no human knew these things. Sometimes I think that's why they're so commonly learned: they are no longer the territory of A Great Man and his disciples, just common knowledge a hair above common sense. Certainly, nobody who tackles these subjects only to be confronted by obnoxious fools with, "How dare you learn only the timeless mathematical truth and ignore the circumstances of its discovery?!"
In one place it says: "The most important things the shuttle group does -- carefully planning the software in advance, writing no code until the design is complete, making no changes without supporting blueprints, keeping a completely accurate record of the code -- are not expensive."
And in another: "on a dollars-per-line basis, it [the group is] among the nation's most expensive software organizations."
Of course, those things are exactly what make it so expensive. It doesn't sound expensive because it's just work. Only, it's a vast amount of work, done by highly-paid professionals. Another bit mentions a change to 6366 lines of code, that was specified by 2500 pages of documentation.
The subject of their efforts is 420,000 lines of code. They've had over 20 years, with a budget currently at 32 million dollars per year. That's enough money for man-weeks for each line of code!
...and they still have bugs.
Fixed prices, subscription quotas, and the threat of withholding (effectively destroying) finished work?
It's complex, inflexible, and requires too much administration and too large a commitment. Can you see yourself signing up for donation subscriptions to support a hundred software projects? Surely you benefit from the efforts of at least that many.
I believe that the best solution is the simplest: just give money to people who have made the stuff you like, don't make them screw around with their own weird variant to make it sound like something other than donation. Don't withhold money from one that is already profitable by the donations of others; huge profits mean that others will be attracted to compete in the area, that's how capitalism allocates resources efficiently over the long term.
If you do that, then people can just get down to the work of figuring out what you want, and making it, knowing from past experience that you can be trusted to make such efforts profitable without coersion.
I left out the part about how, because due to time dilation, he will not be truly dead for several million years at least, the usual inheritance tax was not levied on his estate.
Stalin, while relaxing in a tropical amusement park, picking his nose, was struck by a coconut, prompting a brilliant idea! He immediately dragged his car (he couldn't get any wheel's, because they were patented) to his workshop and made a set of airproof underpants. He considered using charcoal filter's to absorb fart's, but because it would have to be changed, he instead leveraged his recent singularity research to create a small, contained black hole.
To test them, knowing that adult's might be too polite to be honest or too dignified to participate, he gathered together a group of children. The test's were held in a shower to handle any accidents that might result from forced farting. It worked brilliantly. The children tried them in turn's, and not an unpleasant whiff escaped.
For a final test, Stalin tried them on himself and stepped into the shower and strained with all his strength. Unfortunately, his mighty blast destabilized the black hole, causing him and the shower curtain to be sucked into it. At first thinking it all part of the fun, the children were overjoyed at the spectacle.
He was awarded an Ig Nobel Prize posthumously, though this is likely no consolation as he is presumed to live in an eternal hell of contained fart's.
(I swear it sounded like a good idea when I started writing...)
Obviously you can't have young hydrogen-only stars unless the universe was different than it is today.
There are other possibilities.
The universe could be infinite and expanding (everything moving farther apart from each other), with new matter appearing in the voids created. If it expands in the right way, effects of a local event can propagate infinitely into new space (populated with new matter) without ever affecting anything that was a beyond a certain distance from the event at the time it happened.
Or the universe could be finite and self-refreshing on large or small scales. Setting aside the big bang/big crunch idea for the moment, matter and energy could "evaporate" at a rate relative to the gravitational constant in the area, to "condense" in areas with relatively low gravity, in a counter-entropic process that produces fresh hydrogen from worn out iron, neutronium, and um... blackholium?
We don't know the nature of our universe, or its origin, and never will be certain that we do. I don't consider natural history to be real science, any more than human history is. They are both certainly worthy pursuits, but are not built on the truly falsifiable claims which characterize science. I mean, you can assume that the laws of physics that work today applied before any human lived, but you can't test it.
I guess my point is that natural history has to be consistent with both current observation and physics, and we know our physics is incomplete. It's a bit early in the game to go chiseling events that happened billions of years ago onto stone tablets and ridiculing dissenters.
overqualified?
They're using rubbing alcohol and liquid oxygen. Despite the fact that low-concentration hydrogen peroxide might be on your bathroom shelf beside the isopropyl alcohol, they are entirely different chemicals.
Think of how much money goes into car racing. Rocket racing would be an incredible spectacle.
This could easily lead to full funding for the transitional stage of private rocketry before the obvious profit potentials of orbital flight.
They'll never get out the exploits like vaccines and cures if they try to develop it in a closed development group. There are many skilled reverse-engineers in biology. Security through obscurity never works.
Open Source Black Death!
What are they mocking here?
Copyright law or some people's ridiculous mechanistic interpretation of it?
The thing is, there are limits to copyright. The shorter a work, the harder to defend its copyright in court. For instance, it is impossible to copyright a word, phrase, note, or chord. Short poems, like haiku, push the lower bounds, and have quite weak protection: only a very blatant direct copy might infringe on them.
Obviously, these are not legal (or at least not legally relevant) copyrights, and couldn't be enforced.
I know it's all in fun, but I think it would be more satisfying to mock the system using things that would stand up in court.
Experian is a company that sells crack by spamming millions of schoolchildren. MAPS (the Multidisciplinary Association for Psychedelic Studies) has refused to provide their drug-friendly mailing lists because they insist that crack is not a psychedelic drug. However, Experian threatened to take them to court with the argument that if they can include marijuana under their umbrella, then the definition is broad enough to include crack.
Wisely recognizing that both sides are better off not attracting the attention of the courts, MAPS has apparently backed down.
A loss indeed. You can expect many of your peaceful local potheads to become violent criminal crackheads any day now.
I hope this clears up any misconceptions you may have had from the shamefully vague top-level story. I'm a little fuzzy on some of the details myself, but as usual, trying a few likely domain names gave me access to the essentials.
You might be interested in EROS - the Extremely Reliable Operating System, which takes permissions resolution to its logical extreme: the capability system. If something only needs access to one directory and one port, that's all you give it.
Very interesting project.
This just reminded me of a particular Daily Victim.
"In a fit of rage I went over the deep end and cut our apartment's DSL connection!"
The paper is here.
It's more involved than you might think. If you are a sysadmin, this might be important for your job security.
How about a(n empty) "do call" list? Everybody belongs on the "do not call" list for spam and telemarketing.