Can he control the timing of his "yes" signals? If so, this could be made considerably more efficient. Currently it takes on average six inputs and two outputs to specify a letter. But if he can look up either immediately, with a delay, or not at all after hearing a stimulus, then he could signal in Morse code, getting a letter out in (usually) only two or three inputs and the same number of outputs. With better control of his timing, he could signal any of several choices with a single carefully-timed motion.
Prey is unscientific tripe...
on
Prey
·
· Score: 1
Basically, he doesn't know science, the story isn't even realistic, and the horror scenes are all derivative. Evolution doesn't work the way he says it does.
The review contains spoilers, but trust me, you won't miss anything.
Michael Moore has made a great movie about America's culture of guns and fear. It's mostly documentary-style, and covers a lot of ground, including an interview with the head of the NRA.
He also interviews some Canadians. See, Canada has lots of guns, easily available, and very little gun violence. And people don't lock their doors in Canada. A typical interview with a typical Canadian woman (from memory):
"Do you lock your doors at night?"
"No."
"Has anything bad happened?"
"Not really. People have come into my house and vandalized it while I was asleep."
"And you still don't lock your doors?"
"No."
"Is there anything you're afraid of?"
[laughs] "Not really, no."
Quite a contrast from American mindset! Yes, America may be heading for the world of 22 Buttons, but it's quite unnecessary. We let our media scare us. He quoted a very interesting statistic. In some recent time period, gun murders were down 30%, reporting up 600%--or something like that; I clearly remember the 600.
Gray Goo is not a "key part" of Foresight's story. It's not even all that important. You may want to take another look at Foresight--it sounds like you haven't looked at them for several years.
I think you're right about self-replication not being so important. A single desktop factory with fractal converging assembly lines is much easier to program than a mass of individual free-floating assemblers. And in fact, in Nanosystems (written in 1992) Drexler proposes just such a factory.
AP, Portugal, July 4, 1507: The Nina, Pinta, and Santa Maria, used by Cristobal Colon in his discovery of a new route to the Indies, were deliberately destroyed last night, being set afire by their owners.
"They were old, and had rats," explained a Portugese government official. "Our main concern is to save a bit of money."
There has been speculation that England had asked Portugal to destroy the ships as a symbol of England's recent success at colonial enterprises.
Rumors of the impending destruction had been circulating for several months. A few rich investors wanted to use the ships as a museum or hotel, but plans fell through. "It's just too expensive and difficult to keep a fifteen-year-old ship afloat, and not enough people want to spend the money and take the risk of going for a cruise on it," said one potential buyer.
The public has been inexplicably gleeful at the destruction, focusing on the rats and the leaks instead of the historical significance of the vessels. Many who have been following the story made bets on how long the ships would burn once they were set afire. The Nina sank at 1:01 PM, the Pinta and Santa Maria soon after.
Charred bits of wood drifted ashore and soon appeared in the market next to pieces of the True Cross.
I am the coordinator of the Foresight Intellectual Property Reform Project, and I was the one who started this ball rolling last fall. I'd like to clarify a few points.
This project is not about getting patents. It is about preventing and mitigating bad patents. How? By putting prior art where patent examiners will be able to find it easily, and publishing ideas in a way that we can prove when they were published.
The project is not about supporting the patent system. (The title of the Salon article may have caused some confusion on that point.) It is about trying to mitigate the damage caused by bad patents.
The original plan was that Foresight would provide a web site where people could go to publish their ideas in order to keep someone else from patenting them later. We soon realized that publishing ideas in a useful (legally significant) way is not as easy as it sounds, and we found that IP.com was doing almost exactly what we needed to do (for $107 per idea, and aimed at commercial interests). So we talked with them, and sold them on the idea of a much cheaper publication mechanism, for non-commercial ideas only, that could be useful to the Open Source community. Of course they can't give it away; they need to eat, after all. But...
There will be at least two websites. IP.com will offer this service as a product. In addition, the Foresight Institute is planning to get funding and offer this publication service for free for inventions in areas that we care about, including Open Source and nanotechnology.
As long as we have the patent system, we might as well try to deal with its problems. Here are two: Patent examiners have about eight hours to examine prior art. And patent fights tend to cost about US$1,000,000. If you put your idea on a web site or newsgroup somewhere, or use it publicly, chances are good that a patent examiner will not be able to find it in eight hours. So a later, abusive patent is likely to slip through the system. Then, you have to prove in court, years after the fact, that the idea was available to the public before the patent holder claimed to invent it.
The patent examiners actually look at IP.com's database. That's worth a lot. In addition, IP.com has a system set up for proving when the idea was published, and that it was available to the public continuously since that time. They add significant value that we felt would help this project. If you don't want to use it, then don't use it.
Quoting from the court decision: "The evidence before us indicates that the billing process for the electronic stock chart would not actually commence until the client system sent a message to the server system indicating
that the electronic stock chart had been received at the client system. In its brief, Amazon argues that this feature of the CompuServe Trend system amounts to an additional "confirmation
step necessary to complete the ordering process," and that the CompuServe Trend system therefore does not use "single action" technology within the scope of the claims in the '411
patent."
If I understand this, Amazon says that Trend doesn't count because the client's computer had to send something back to the server after the click that ordered the chart. Now... when does a modern one-click shopping system check the user's cookie to see who clicked on their one-click button? Doesn't it retrieve the cookie after receiving the click?
Does this mean that Amazon has argued that no actual one-click system using cookies is covered by their patent???
If this thread is interesting, don't miss Shadow Syndromes by John J. Ratey and Catherine Johnson, which describes mild versions of several disorders such as autism and OCD.
For info on how another mental disability can improve intelligence and creativity, check out The Gift of Dyslexia by Ronald D. Davis. (Oh, you want a URL? Try the Davis web site or stuff I wrote about it.)
The software seems to pick out the most unusual words in a page. Typos can get quite unusual. One of their papers gives an example that uses "peroperties" as an index word. On the target page, it's clearly a typo for "properties". If the authors of that page ever bothered to spell-check it, that word would go away, and the paper would be that much harder to find.
Octavia Butler: Can't praise her highly enough. Writes several genres. For SF, try the Xenogenesis trilogy (Dawn, Adulthood Rites, Imago). Human-alien contact with several twists.
Sheri S. Tepper: A bit preachy, but very interesting worlds and characters. Start with the Awakeners (Northshore/Southshore) or Raising the Stones.
Samuel R. Delany: Mind-expanding in several dimensions at once. Preview for sex before giving to kids. Start with Triton or Stars In My Pocket Like Grains Of Sand.
Henry Kuttner, aka Lewis Padgett: Generally excellent. Read his short story collections, eg. The Best Of Henry Kuttner (don't miss Mimsy Were the Borogoves). Also a novelette, Vintage Season.
Alfred Bester: The Demolished Man and The Stars My Destination (original title: Tyger Tyger). TDM is the reason the head of Psi Corps in B5 is named Bester.
John Varley: Consistently excellent novels and short stories. Don't miss The Persistence Of Vision.
Thomas M. Disch: Fun With Your New Head, 102 H-bombs (especially the title story).
Tanith Lee: The Silver Metal Lover (warning--bites deep).
Edgar Pangborn: The Freshman Angle, a hard-to-find short story in Ten Tomorrows.
You raise some good questions. First, about medicine, you're right that we can't build something we can't understand. But... computers millions of times faster than today's will make it somewhat easier to understand some complex systems, such as our bodies. And the ability to build any research tool we can design will be a huge help to medical researchers. Imagine if you had a matter compiler on your desk, and could have the latest subcellular probe an hour after you downloaded it off the Web. Or imagine if you could safely probe a million neurons at once in a living human brain. Medical research will accelerate hugely once nanotech arrives.
As to social issues: there's a wild card that people aren't talking about much yet. Artificial intelligence will be a major influence on the way we run society. Surveillance will get a lot easier. Propaganda (advertising) will become more powerful. If the AIs become a lot smarter than we are, they may even comprehend us enough to do social engineering. Will they be working for our benefit? How could we tell? Nanotech by itself won't give us superhuman AI, but it will give us computers that can simulate every neuron in our brains a million times faster. That's a good place to start.
I sent this to the "comments" address at the bottom of the column. I kept it a lot milder than I felt...
I was concerned by several comments in Moira Gunn's recent article on Lucas' attempts to stop Internet piracy of Episode 1. I began to worry when she wrote, "They started out by warning some 700 Internet Service Providers they would be held responsible for anyone offering bootleg copies on their services." Last I heard, ISPs are not responsible for content that they don't edit, especially if they're not aware of it. Has this changed? I hope not, because that would chill a lot of socially valuable communication--ISPs would have to become much too restrictive, rather than simply allowing the law to deal with the people who actually commit the crime. But if ISPs are not in fact responsible, then what Lucas did was simply dishonest bullying.
My worry deepened at the end of the article, where she proposed, "you can really make it uncomfortable for people who do irritating things on the Internet.... Why not have a `Cyber Patrol` which is constantly vigilant, watching the World Wide Web?" Well, there are several groups I don't trust, such as the NSA (remember the Clipper Chip fiasco?) and the Religious Right, who would be delighted to build such a "Cyber Patrol." Let's not give them too many ideas. Her mention of combating loss of privacy is ironic--a Cyber Patrol would likely "expose" much legitimate business that should not be taken out of context, such as speech that is normal for California but violates the community standards of the Bible Belt. Let's not forget that people have been arrested and dragged cross-country--and convicted!--for pursuing legitimate, legal, normal business that some minor official in Tennessee didn't happen to like.
Finally, Ms. Gunn writes, "We can stop most [Internet information] with a little determination." Is this really what we want? Do we want the ability to stop information about civil rights violations around the globe? (Remember Tianeman Square?) Do we want the ability to silence debate about dangerous organizations? (The Scientologists are doing a pretty good job of that now.) Do we want to give special-interest groups the ability to legislate what can and can't be said between adults? (The Communications Decency Act was an unconstitutional travesty--but it's not dead yet!)
Governments and lawyers are quite appropriate when engaged in preventing actual theft. But covering the whole Web in a blanket of lawyers will do far more harm than good. Imagine how different our society would be if every telephone conversation and postal letter were monitored as a matter of course. It's hard to see a dividing line between such monitoring and Ms. Gunn's proposal. It looks to me more like a slippery slope.
As a side note, Ms. Gunn claims that Lucas' lawyers "stopped the great bulk of the bootleggers." How confident are you in this claim? The article gave no evidence whatever.
You spit blood over dialing a new prefix... what if, for no good reason, you had to tell everyone that your mailing address changed? And the post office would refuse to deliver to the old address? AND if they investigated and decided that you were running a business through your mailbox, they would give your "business" (= home) address to anyone who asked? And it will cost small businesses several billion dollars. This is not a joke. The US post office has already made this rule, quietly, despite 8,000 comments against it. Check out http://www.postalwatch.org for more details...
I believe this is common practice in avionics (flight computers). You run three computers, and if one of them produces an output different from the other two, you know you should ignore it. Seems to me many years ago I read a news article describing such a system in the Space Shuttle.
Perhaps the difference is that they specify a "test" and "reference" system rather than three systems that are equally under suspicion.
The patent has 23 claims. Some of them are quite specific: e.g. using a binary search to find the problem. But claim 1 is incredibly general.
Slashdot Mirror
Can he control the timing of his "yes" signals? If so, this could be made considerably more efficient. Currently it takes on average six inputs and two outputs to specify a letter. But if he can look up either immediately, with a delay, or not at all after hearing a stimulus, then he could signal in Morse code, getting a letter out in (usually) only two or three inputs and the same number of outputs. With better control of his timing, he could signal any of several choices with a single carefully-timed motion.
See *here* for a review.
Basically, he doesn't know science, the story isn't even realistic, and the horror scenes are all derivative. Evolution doesn't work the way he says it does.
The review contains spoilers, but trust me, you won't miss anything.
Chris
He also interviews some Canadians. See, Canada has lots of guns, easily available, and very little gun violence. And people don't lock their doors in Canada. A typical interview with a typical Canadian woman (from memory):
"Do you lock your doors at night?"
"No."
"Has anything bad happened?"
"Not really. People have come into my house and vandalized it while I was asleep."
"And you still don't lock your doors?"
"No."
"Is there anything you're afraid of?"
[laughs] "Not really, no."
Quite a contrast from American mindset! Yes, America may be heading for the world of 22 Buttons, but it's quite unnecessary. We let our media scare us. He quoted a very interesting statistic. In some recent time period, gun murders were down 30%, reporting up 600%--or something like that; I clearly remember the 600.
Chris
Gray Goo is not a "key part" of Foresight's story. It's not even all that important. You may want to take another look at Foresight--it sounds like you haven't looked at them for several years.
I think you're right about self-replication not being so important. A single desktop factory with fractal converging assembly lines is much easier to program than a mass of individual free-floating assemblers. And in fact, in Nanosystems (written in 1992) Drexler proposes just such a factory.
Chris
Pioneering ships destroyed
AP, Portugal, July 4, 1507: The Nina, Pinta, and Santa Maria, used by Cristobal Colon in his discovery of a new route to the Indies, were deliberately destroyed last night, being set afire by their owners.
"They were old, and had rats," explained a Portugese government official. "Our main concern is to save a bit of money."
There has been speculation that England had asked Portugal to destroy the ships as a symbol of England's recent success at colonial enterprises.
Rumors of the impending destruction had been circulating for several months. A few rich investors wanted to use the ships as a museum or hotel, but plans fell through. "It's just too expensive and difficult to keep a fifteen-year-old ship afloat, and not enough people want to spend the money and take the risk of going for a cruise on it," said one potential buyer.
The public has been inexplicably gleeful at the destruction, focusing on the rats and the leaks instead of the historical significance of the vessels. Many who have been following the story made bets on how long the ships would burn once they were set afire. The Nina sank at 1:01 PM, the Pinta and Santa Maria soon after.
Charred bits of wood drifted ashore and soon appeared in the market next to pieces of the True Cross.
Chris Phoenix
This project is not about getting patents. It is about preventing and mitigating bad patents. How? By putting prior art where patent examiners will be able to find it easily, and publishing ideas in a way that we can prove when they were published.
The project is not about supporting the patent system. (The title of the Salon article may have caused some confusion on that point.) It is about trying to mitigate the damage caused by bad patents.
The original plan was that Foresight would provide a web site where people could go to publish their ideas in order to keep someone else from patenting them later. We soon realized that publishing ideas in a useful (legally significant) way is not as easy as it sounds, and we found that IP.com was doing almost exactly what we needed to do (for $107 per idea, and aimed at commercial interests). So we talked with them, and sold them on the idea of a much cheaper publication mechanism, for non-commercial ideas only, that could be useful to the Open Source community. Of course they can't give it away; they need to eat, after all. But...
There will be at least two websites. IP.com will offer this service as a product. In addition, the Foresight Institute is planning to get funding and offer this publication service for free for inventions in areas that we care about, including Open Source and nanotechnology.
As long as we have the patent system, we might as well try to deal with its problems. Here are two: Patent examiners have about eight hours to examine prior art. And patent fights tend to cost about US$1,000,000. If you put your idea on a web site or newsgroup somewhere, or use it publicly, chances are good that a patent examiner will not be able to find it in eight hours. So a later, abusive patent is likely to slip through the system. Then, you have to prove in court, years after the fact, that the idea was available to the public before the patent holder claimed to invent it.
The patent examiners actually look at IP.com's database. That's worth a lot. In addition, IP.com has a system set up for proving when the idea was published, and that it was available to the public continuously since that time. They add significant value that we felt would help this project. If you don't want to use it, then don't use it.
Chris Phoenix
If I understand this, Amazon says that Trend doesn't count because the client's computer had to send something back to the server after the click that ordered the chart. Now... when does a modern one-click shopping system check the user's cookie to see who clicked on their one-click button? Doesn't it retrieve the cookie after receiving the click?
Does this mean that Amazon has argued that no actual one-click system using cookies is covered by their patent???
Chris
For info on how another mental disability can improve intelligence and creativity, check out The Gift of Dyslexia by Ronald D. Davis. (Oh, you want a URL? Try the Davis web site or stuff I wrote about it.)
The software seems to pick out the most unusual words in a page. Typos can get quite unusual. One of their papers gives an example that uses "peroperties" as an index word. On the target page, it's clearly a typo for "properties". If the authors of that page ever bothered to spell-check it, that word would go away, and the paper would be that much harder to find.
(I've already sent them an email about this.)
Chris
Sheri S. Tepper: A bit preachy, but very interesting worlds and characters. Start with the Awakeners (Northshore/Southshore) or Raising the Stones.
Samuel R. Delany: Mind-expanding in several dimensions at once. Preview for sex before giving to kids. Start with Triton or Stars In My Pocket Like Grains Of Sand.
Henry Kuttner, aka Lewis Padgett: Generally excellent. Read his short story collections, eg. The Best Of Henry Kuttner (don't miss Mimsy Were the Borogoves). Also a novelette, Vintage Season.
Alfred Bester: The Demolished Man and The Stars My Destination (original title: Tyger Tyger). TDM is the reason the head of Psi Corps in B5 is named Bester.
John Varley: Consistently excellent novels and short stories. Don't miss The Persistence Of Vision.
Thomas M. Disch: Fun With Your New Head, 102 H-bombs (especially the title story).
Tanith Lee: The Silver Metal Lover (warning--bites deep).
Edgar Pangborn: The Freshman Angle, a hard-to-find short story in Ten Tomorrows.
You raise some good questions. First, about medicine, you're right that we can't build something we can't understand. But... computers millions of times faster than today's will make it somewhat easier to understand some complex systems, such as our bodies. And the ability to build any research tool we can design will be a huge help to medical researchers. Imagine if you had a matter compiler on your desk, and could have the latest subcellular probe an hour after you downloaded it off the Web. Or imagine if you could safely probe a million neurons at once in a living human brain. Medical research will accelerate hugely once nanotech arrives.
As to social issues: there's a wild card that people aren't talking about much yet. Artificial intelligence will be a major influence on the way we run society. Surveillance will get a lot easier. Propaganda (advertising) will become more powerful. If the AIs become a lot smarter than we are, they may even comprehend us enough to do social engineering. Will they be working for our benefit? How could we tell? Nanotech by itself won't give us superhuman AI, but it will give us computers that can simulate every neuron in our brains a million times faster. That's a good place to start.
Chris
I sent this to the "comments" address at the bottom of the column. I kept it a lot milder than I felt...
I was concerned by several comments in Moira Gunn's recent article on Lucas' attempts to stop Internet piracy of Episode 1. I began to worry when she wrote, "They started out by warning some 700 Internet Service Providers they would be held responsible for anyone offering bootleg copies on their services." Last I heard, ISPs are not responsible for content that they don't edit, especially if they're not aware of it. Has this changed? I hope not, because that would chill a lot of socially valuable communication--ISPs would have to become much too restrictive, rather than simply allowing the law to deal with the people who actually commit the crime. But if ISPs are not in fact responsible, then what Lucas did was simply dishonest bullying.
My worry deepened at the end of the article, where she proposed, "you can really make it uncomfortable for people who do irritating things on the Internet.... Why not have a `Cyber Patrol` which is constantly vigilant, watching the World Wide Web?" Well, there are several groups I don't trust, such as the NSA (remember the Clipper Chip fiasco?) and the Religious Right, who would be delighted to build such a "Cyber Patrol." Let's not give them too many ideas. Her mention of combating loss of privacy is ironic--a Cyber Patrol would likely "expose" much legitimate business that should not be taken out of context, such as speech that is normal for California but violates the community standards of the Bible Belt. Let's not forget that people have been arrested and dragged cross-country--and convicted!--for pursuing legitimate, legal, normal business that some minor official in Tennessee didn't happen to like.
Finally, Ms. Gunn writes, "We can stop most [Internet information] with a little determination." Is this really what we want? Do we want the ability to stop information about civil rights violations around the globe? (Remember Tianeman Square?) Do we want the ability to silence debate about dangerous organizations? (The Scientologists are doing a pretty good job of that now.) Do we want to give special-interest groups the ability to legislate what can and can't be said between adults? (The Communications Decency Act was an unconstitutional travesty--but it's not dead yet!)
Governments and lawyers are quite appropriate when engaged in preventing actual theft. But covering the whole Web in a blanket of lawyers will do far more harm than good. Imagine how different our society would be if every telephone conversation and postal letter were monitored as a matter of course. It's hard to see a dividing line between such monitoring and Ms. Gunn's proposal. It looks to me more like a slippery slope.
As a side note, Ms. Gunn claims that Lucas' lawyers "stopped the great bulk of the bootleggers." How confident are you in this claim? The article gave no evidence whatever.
Chris
You spit blood over dialing a new prefix... what if, for no good reason, you had to tell everyone that your mailing address changed? And the post office would refuse to deliver to the old address? AND if they investigated and decided that you were running a business through your mailbox, they would give your "business" (= home) address to anyone who asked? And it will cost small businesses several billion dollars.
This is not a joke. The US post office has already made this rule, quietly, despite 8,000 comments against it. Check out http://www.postalwatch.org for more details...
Chris
I believe this is common practice in avionics (flight computers). You run three computers, and if one of them produces an output different from the other two, you know you should ignore it. Seems to me many years ago I read a news article describing such a system in the Space Shuttle.
Perhaps the difference is that they specify a "test" and "reference" system rather than three systems that are equally under suspicion.
The patent has 23 claims. Some of them are quite specific: e.g. using a binary search to find the problem. But claim 1 is incredibly general.
Chris