One of the main reasons the 9/11 operation was so successful was that it was very closely held. Some of the actual participants didn't know the plan until the day of the event. It didn't take much in the way of resources; a few men willing to die and about $250,000.
In a just society every area of activity needs at least two strong opposing forces, each of which prevents the other from running amok. In economics the two are capital and regulation.
No, communism used to fulfill that function.
From 1917 to 1980 or so, communism was a serious political threat to unrestrained capitalism. If capitalism didn't deliver a higher standard of living for the common man in a democracy, it might be voted out. That pushed wages up. Capitalists were scared of the competition. There was a real worry that communism might deliver the goods. It did in some areas. The USSR worked to provide jobs, housing, and medical care for everyone. None were all that great, but they beat being homeless.
In response to this, the US made a big political point of maintaining a continuously rising standard of living. That's what the "American Way of Life" was all about. Through at least the mid-1970s, that position was politically mainstream and supported by both parties. Labor unions helped to keep the pressure on and wages up, both for union and non-union members.
The USSR started to tank when their own people stopped believing in the dream. By 1980, the ideology was an empty shell. It took another decade for the system to run down, but it did.
After that, capitalism was the only game in town. There was no reason to pay workers more than they were economically worth. Money then moved from workers to owners. Unions were crushed.
Without ideological competition, we got to where we are now.
This is a ripe time for a charismatic leader to tell them what to think, and gin up some will to act decisively. Its also notable that heading into presidential elections none of the candidates are that person. Obama is out there trying to be and its not working.
Very true. Obama ran for President like he was going to be FDR, and he turned out to be Herbert Hoover. For the first two years, he had a big majority in the House and Senate to work with, and still didn't accomplish much.
The Republican candidates are at best mediocre, and at worst nuts. The ones that could probably do an adequate job are out of the race. What the Republicans have right now are people who say "tax cuts for the rich" to their contributors, and "God, guns and gays" to the rabble. (Listen to Herman Cain's speech to the Values Voter Conference for an example.)
The trouble is that charismatic leaders historically don't work out all that well. That's how you get dictators.
The really effective presidents since WWII, Truman, Eisenhower, Johnson, and maybe Bush I, were more about competence than charisma.
We don't need a leader. We need an agenda. A few key points, and a general direction.
I remember the '60s, they sucked the first time. And the anti-war protests of the '60s tended to be much more focused than this crap.
That's because the main agenda was to avoid being drafted. The anti-war movement was about self-interest. There hasn't been a serious anti-war movement in the US since the military went all volunteer.
This time, the agenda is to avoid being laid off and broke. That, too, has self-interest driving it.
That's what Karl Rove did for decades. There's a classic picture of Karl Rove with a 12" reel of computer tape. He was able to turn demographic and polling data into information on how, when, and where to tell people what they wanted to hear.
You mean like Lynas mining's Mt Weld in Australia?
Right. And Mountain Pass, California.
The big problem with rare earth mining (and gold mining) is finding a place to dump the wastes. Some rare earth mines are environmental disaster areas. Most are in the middle of nowhere. Mountain Pass CA is adjacent to I-15, east of Barstow. They have a unique solution - they're building a pipeline to Nevada so they can dump the sludge onto low value real estate in a red state.
The ominous part of that memo is "The cyclone of innovation is increasingly moving off the web onto iOS and other closed platforms." By which they mean iPhone and Android "apps". "Apps" are not a very good environment, and many of them are just web pages with delusions of grandeur. But that they have a payment model, DRM, and give the app distributor absolute control.
It's all about screwing the end user. "You're the product, not the customer".
As a language, Dash looks mediocre, as the article points out. "Optional typing", an idea that started with Visual Basic is usually a lose in language design. Statically typed languages have been successful, and dynamically typed languages have been successful, but optional typing is usually an afterthought bolted on to increase performance at the cost of programmer confusion. There's a typed Python variant, for example; PyPy is written in it. It's rarely put in a language from the beginning.
A few languages have tried a form of soft typing, where you have, essentially,"integer", "real", "boolean", etc., and arrays of same, plus "object". That way you get efficient code for machine arithmetic, which means you can do codecs and graphics in the language. Objects have to be dispatched anyway, so a performance penalty there isn't so severe.
Having worked in the classified world (pre 9/11), it was surprising how little military information was classified. The front-line military view of secrecy is that secrecy is a short-term thing. "Where the ship was last week is unclassified. Where the ship was yesterday is confidential. Where the ship is now is secret. Where the ship will be tomorrow is top secret." Sooner or later, if it matters, the enemy will find out what you're up to. Preferably when your attack hits them.
On the other hand, what your troops, ships and planes can do is generally well known. Too many people have to know. Secret capabilities do exist, but, again, they're time-sensitive. Eventually you have to use the secret weapon, after which it's no longer secret.
Vulnerabilities are more of a problem. The U.S. Army tried to keep secret the vulnerable spots on a M-1 Abrams tank. But once Iraqi insurgents had found the places on the turret ring to aim at, trying to suppress the pictures of the damage was sort of stupid.
When planning proposals, we estimated that running a project at SECRET doubled the cost, and running at TOP SECRET quadrupled it. (The clearance process takes many months, the physical security is expensive and slows you down, and worst of all, the people who spend too much time in classified tanks get out of touch technically.)
The intel community was willing to pay that price - the military, not so much.
Sounding rockets in this size have been around for a while. The first one was the Aerobee sounding rocket in 1947. It reached 117500m. One of the smaller Aerobee variants of the 1950s was about this size. There have been many small sounding rockets over the decades; the UK and Australia launched a lot of them.
Some intellectual property lawyer from UC Berkeley has been quoted as saying that Berkeley should have patented TCP/IP. (Can anyone find this quote). UC Berkeley didn't either design TCP/IP or do the first TCP/IP implementation. They did the fourth or fifth. (BBN, Phil Karn, Dave Mills, and 3COM all had earlier implementations.) They didn't even have the first UNIX implementation; that was 3COM's UNET. What BSD had was government funding to give their implementation away. UNET was about $3K per machine.
The trouble with "malicious URL scanners" that look for malware is that unless they're real-time, they're too late. The lifetime of bad sites is now often measured in hours.
Still, continually detecting the bad guys and beating on them does have effect. Major services have to do it, or they get pwned.
We do some tracking of major sites being exploited by phishers, There are only 29 sites on the list today, one of the shortest lists we've had in years. It's been as high as 140. The URL-shortening sites get hit daily, and kick most of the bad guys off within hours. The free-hosting sites get hit too, and most of them are good about kicking the phishers off. (Piczo and webs.com are not too good at this. t35 has gotten much better. Google has a big problem with people using documents and spreadsheets as phishing sites.)
This makes Google's browser hostile code. It should not be allowed through corporate firewalls. On the browser front, progress has been made by giving parts of the browser that run external code less privilege. Sandboxing Flash and Acrobat Reader is progress. Mozilla's dividing of add-ons into a non privileged content script and a somewhat more privileged add-on code is progress. Putting an equivalent of Back Orifice into a browser is not.
The announcement says: the technology right now is limited so that permission must be granted each time remote administration is activated. How long will that last? Could be changed silently by a forced update? What if law enforcement wants to use it? Does the remote session run through a Google server? (The protocol is apparently based on Google Talk, which does.) How else do they get two clients behind DHCP routers talking to each other? Is the connection encrypted? Is it encrypted end to end, or is the server in a position to mount a man in the middle attack? Does Google commit contractually to not accessing your machine, or is there an EULA that says they can do that whenever they want to?
If you want remote desktop access in the corporate environment, there are management tools for that. They're usually locked down tightly, since they're inherently a security risk.
Right now, we have two main models for storage - files and a flat address space. Neither is well suited to flash memory, let alone something like memristors.
There are other architectures. Burroughs machines from 1960 onward had memory addressing that worked like pathnames. Think of memory addresses as being hierarchical, like "process22/object21/array4[5]". Objects were paged in and out, but were not persistent. The IBM System/38 went further and made such objects persistent. However, such architectures are not compatible with C programming, which assumes that addresses are numbers.
Flash is usually treated as disk, even though it has an access time that's faster than the time the OS takes to grind through the file system code. Flash is much more of a random access device, but it's seldom supported as one. One possibility would be to support flash memory as a name/value tuple store, like the "non-SQL" databases.
PC-type architectures still don't have a channel architecture (like IBM mainframes since 1967), where a user program can have non-privileged access to a device without the OS being in the middle. (Channel support means a process can talk to a device directly by sending messages, messages which the device can securely associate with the sending process and its privileges. The device has to maintain security and has the information to do so.)
So smart devices that do lots of little fast transactions, like a tuple store or an object store, have to be mediated through the OS. So most OSs are still reading big blocks from flash, then caching them in RAM. That's already limiting, and memristors make it more so.
The success of C and the UNIX model (which even Windows NT and its successor follow) is based on a vanilla hardware model. That's holding computing back.
Obviously they don't have a "magnetic beam" problem. They might have some electrical wiring problem. Those are easy to find.
Only once have I seen a real "magnet problem". I was trying to get a flux-gate compass to work in a mobile robot at Stanford, and was getting bogus results. So I got an ordinary needle compass, and observed that it didn't point north. I walked around with the compass plotting directions, and the center of the problem was a small building about a block away. When I went into the building, I saw "High Magnetic Field" warning signs, and found out there was a superconducting magnet in there. A big one. Even there, it was only one gauss just outside the lab.
The E-Cat is deceptively simple: hydrogen is passed over a special catalyst based on nickel in a container about a litre in size, and enough heat is produced to boil water. A demonstration in January appeared to show a several kilowatts of output from a four hundred watt input.
From the picture, which shows tubes marked "H" and "O2", this looks like a system for converting hydrogen and oxygen into water and heat with a catalyst. That's a routine technology; after all, you can burn H and O2; a catalyst just lets you do it at a lower temperature. This doesn't appear to be a closed system; for that they'd have to crack the water produced (they get steam out) back into H and O2.
How are the volumes of H and O2 going in measured? How is the heat coming out measured? There's some detail on NyTeknik but not much. It's not clear why there's 30Kg of lead in the thing. I wonder what's inside that lead. It's supposedly not producing any high-energy particles. If it did, that would be interesting. They don't seem to have radiation detectors around.
(Some years ago, when the Pons- Fleischmann cold fusion flap was underway, I went to a talk at Stanford by some physicists who were trying to reproduce the experiment. They'd started out with radiation detectors and alarms surrounding the apparatus in case it suddenly produced dangerous amounts of radiation. After a while, it was clear that nothing dramatic was going to happen. They were trying to measure neutron output, and background radiation was more than whatever the cold fusion apparatus was putting out. They finally put it in a big box of lead cubes to get rid of background neutrons, and still couldn't measure any neutrons coming out. General feeling of exasperation after weeks of work.)
All these short tests of an hour or day are suspicious. If this thing really worked, they could set it on a glass table (so observers could check for external connections) and run it for a month. The short runs hint that some consumable is being used up.
After half a century of building big rockets, we now know that they don't work very well. Half a century ago, they were use-once-and-throw-away devices, and they still are. Payloads are still tiny compared to the launch weight, even for the Shuttle. Compare the figures for jet aircraft, which can be half payload.
Reliability is still lousy, too. This is because so much weight reduction is required just to get the things off the ground that they don't have adequate safety margins. About 10-20% of satellite launches still fail, almost half a century after the first one. That number isn't improving, either; in fact, it was a little better in the 1970s. There have only been a few hundred Shuttle flights, and it's crashed once. (Update since I wrote this in 2002: twice). Commercial aircraft flights, by comparison, fail a few times per year, out of millions of flights.
Half a century in aviation took us from the Wright Brothers Flyer to the B-52. Half a century in rocketry took us from the Atlas I to the Atlas V. There's been little progress in launch vehicles since the 1960s. All the major launch systems were created decades ago. So chemical fuels just don't have the power-to-weight ratio for useful space travel. People knew this in the Orion nuclear rocket days; it's a straightforward calculation. It's unfortunate that an Orion wasn't launched once or twice, just to demonstrate that nuclear propulsion is possible.
The bullying they're concerned about mostly involves gay kids. From the paper: "Jamey Rodemeyer, a 14-year old boy from Williamsville, near Buffalo, NY, took his life after what his parents claim was yeas of bullying because of struggles with his sexuality."
"Tyler Clementi, an 18-year-old Rutgers student,
committed suicide by jumping off the George Washington Bridge after two classmates secretly taped him during a sexual encounter with a man and broadcast it over the web."
One of the papers cited is "LGBT and Allied Youth Responses to
Cyberbullying: Policy Implications" On page 128 (the paper starts at journal page 115), after discussing how big a problem this is for gay youth, they remark "This finding somewhat contradicts a study by Smith, et al. (2008) on cyber-
bullying in secondary school, which did not factor the characteristics of sexual
identity and gender identity into the equation. Instead, they found that secondary
school students, in general, recommended as their best coping strategies in coun-
tering cyberbullying both blocking and avoiding messages, and telling someone
when they were being cyberbullied." The key point here is that relatively simple avoidance strategies work for hetrosexual kids, while gay kids have much worse problems. They often can't get help from their parents. "Family rejection is often more feared than victimization or harassment.... a significant percent of LGBT youth are forced to leave home
once their sexual or gender identity is questioned by family members, and approx-
imately 20-40% of all homeless youth are lesbian, gay, bisexual, or transgender".
So this really isn't a free speech problem at all. It's a parenting problem.
It's just that immutability is default, and you have to sprinkle "mutable" in your data structures where you actually need it.
That's a related idea to functional programming - single-assignment languages. The idea is that all (or at least most) names are "const" - they can only be assigned a value once. This works a lot like functional programming, but there's not so much nesting, and most things have names, rather than being anonymous within some nested expression.
Pure functional languages force programs into a tree form, with one output. A single assignment language forces programs into directed acyclic graph form, with multiple outputs, but no loops in the graph. (At least if you have some way to get multiple values returned from a function.)
Many newer "functional" languages are really single-assignment languages. This avoids getting buried in deeply nested parentheses, as with LISP.
Python is similarly terse, and isn't statically typed. So this isn't about static typing. It's another functional programming fan.
Functional programming is a good fit to a certain class of algorithms. For the 65 programmers of a trading house, it makes sense.
Functional programming has its downsides. It tends to result in heavily nested code. It's hard to fan out results, so programs tend to be trees with a single result. Persistent state and I/O don't fit well with the functional model.
OCaml has some features that ought to be in more languages, like Djykstra's "guards" for choosing multiple alternatives.
Microsoft does it again. First they killed PlaysForSure, with its DRM, and now Zune,with its own incompatible DRM.
As I've pointed out before, the lifetime of DRM systems seems to be about five years. At the end of life, users tend to lose content, although sometimes there's a migration path.
Google Earth is really a standalone Windows application. (Remember those?) Google didn't develop it; it was from a company called Keyhole, which sold it as a service for years before Google bought them out. I had a Keyhole account back in 2003. NVidia had a promotional deal; it was cheaper with an NVidia card.
You could fly along a route in Keyhole, so a "helicopter view" isn't exactly new.
Look up the definition of cold shutdown. It doesn't matter if reactor containment is breached.
Not in this case. Here, "cold shutdown" has been redefined somewhat, to "below boiling if we can keep cooling water going in." Normally, in a cold shutdown, you can take the lid off the reactor, look inside, and replace fuel rods. They're a long way from that point.
More like: "the molten corium has burrowed deep enough to be cooled by groundwater and we are only reading 90 degrees at the twisted, melted reactor because the radioactive steam coming from below ground is dissipating the heat"
But not that bad, either. These reactors were built on bedrock. That placed them lower than would have been desirable for flood protection, but if they leak, they leak sideways, not down. There's been plenty of sideways leakage, but by now most of that water is being collected. There's now a cleaning plant in place to run the water through zeolites and catch the radioactive salts and solids. (Water itself doesn't become radioactive from exposure to gamma radiation; the longest lived radioactive isotope of oxygen has a half-life of 122 seconds.)
Now they have to figure out how to do the tough job - safely dismantling the radioactive mess in the melted core into small bits for disposal. That may take decades.
Slashdot Mirror
Oh, not again.
One of the main reasons the 9/11 operation was so successful was that it was very closely held. Some of the actual participants didn't know the plan until the day of the event. It didn't take much in the way of resources; a few men willing to die and about $250,000.
In a just society every area of activity needs at least two strong opposing forces, each of which prevents the other from running amok. In economics the two are capital and regulation.
No, communism used to fulfill that function.
From 1917 to 1980 or so, communism was a serious political threat to unrestrained capitalism. If capitalism didn't deliver a higher standard of living for the common man in a democracy, it might be voted out. That pushed wages up. Capitalists were scared of the competition. There was a real worry that communism might deliver the goods. It did in some areas. The USSR worked to provide jobs, housing, and medical care for everyone. None were all that great, but they beat being homeless.
In response to this, the US made a big political point of maintaining a continuously rising standard of living. That's what the "American Way of Life" was all about. Through at least the mid-1970s, that position was politically mainstream and supported by both parties. Labor unions helped to keep the pressure on and wages up, both for union and non-union members.
The USSR started to tank when their own people stopped believing in the dream. By 1980, the ideology was an empty shell. It took another decade for the system to run down, but it did.
After that, capitalism was the only game in town. There was no reason to pay workers more than they were economically worth. Money then moved from workers to owners. Unions were crushed.
Without ideological competition, we got to where we are now.
This is a ripe time for a charismatic leader to tell them what to think, and gin up some will to act decisively. Its also notable that heading into presidential elections none of the candidates are that person. Obama is out there trying to be and its not working.
Very true. Obama ran for President like he was going to be FDR, and he turned out to be Herbert Hoover. For the first two years, he had a big majority in the House and Senate to work with, and still didn't accomplish much.
The Republican candidates are at best mediocre, and at worst nuts. The ones that could probably do an adequate job are out of the race. What the Republicans have right now are people who say "tax cuts for the rich" to their contributors, and "God, guns and gays" to the rabble. (Listen to Herman Cain's speech to the Values Voter Conference for an example.)
The trouble is that charismatic leaders historically don't work out all that well. That's how you get dictators. The really effective presidents since WWII, Truman, Eisenhower, Johnson, and maybe Bush I, were more about competence than charisma.
We don't need a leader. We need an agenda. A few key points, and a general direction.
I remember the '60s, they sucked the first time. And the anti-war protests of the '60s tended to be much more focused than this crap.
That's because the main agenda was to avoid being drafted. The anti-war movement was about self-interest. There hasn't been a serious anti-war movement in the US since the military went all volunteer.
This time, the agenda is to avoid being laid off and broke. That, too, has self-interest driving it.
How many "Google+" users are just Gmail users who were migrated?
Remember Yahoo 360? Me either.
Google does really well at search and search ads. Everything else they try, not so much.
That's what Karl Rove did for decades. There's a classic picture of Karl Rove with a 12" reel of computer tape. He was able to turn demographic and polling data into information on how, when, and where to tell people what they wanted to hear.
You mean like Lynas mining's Mt Weld in Australia?
Right. And Mountain Pass, California.
The big problem with rare earth mining (and gold mining) is finding a place to dump the wastes. Some rare earth mines are environmental disaster areas. Most are in the middle of nowhere. Mountain Pass CA is adjacent to I-15, east of Barstow. They have a unique solution - they're building a pipeline to Nevada so they can dump the sludge onto low value real estate in a red state.
The ominous part of that memo is "The cyclone of innovation is increasingly moving off the web onto iOS and other closed platforms." By which they mean iPhone and Android "apps". "Apps" are not a very good environment, and many of them are just web pages with delusions of grandeur. But that they have a payment model, DRM, and give the app distributor absolute control.
It's all about screwing the end user. "You're the product, not the customer".
As a language, Dash looks mediocre, as the article points out. "Optional typing", an idea that started with Visual Basic is usually a lose in language design. Statically typed languages have been successful, and dynamically typed languages have been successful, but optional typing is usually an afterthought bolted on to increase performance at the cost of programmer confusion. There's a typed Python variant, for example; PyPy is written in it. It's rarely put in a language from the beginning.
A few languages have tried a form of soft typing, where you have, essentially,"integer", "real", "boolean", etc., and arrays of same, plus "object". That way you get efficient code for machine arithmetic, which means you can do codecs and graphics in the language. Objects have to be dispatched anyway, so a performance penalty there isn't so severe.
That's nice, but scandium has only a few minor uses. A find of high-quality neodymium or europium ore would be much more interesting.
Having worked in the classified world (pre 9/11), it was surprising how little military information was classified. The front-line military view of secrecy is that secrecy is a short-term thing. "Where the ship was last week is unclassified. Where the ship was yesterday is confidential. Where the ship is now is secret. Where the ship will be tomorrow is top secret." Sooner or later, if it matters, the enemy will find out what you're up to. Preferably when your attack hits them.
On the other hand, what your troops, ships and planes can do is generally well known. Too many people have to know. Secret capabilities do exist, but, again, they're time-sensitive. Eventually you have to use the secret weapon, after which it's no longer secret.
Vulnerabilities are more of a problem. The U.S. Army tried to keep secret the vulnerable spots on a M-1 Abrams tank. But once Iraqi insurgents had found the places on the turret ring to aim at, trying to suppress the pictures of the damage was sort of stupid.
When planning proposals, we estimated that running a project at SECRET doubled the cost, and running at TOP SECRET quadrupled it. (The clearance process takes many months, the physical security is expensive and slows you down, and worst of all, the people who spend too much time in classified tanks get out of touch technically.) The intel community was willing to pay that price - the military, not so much.
Sounding rockets in this size have been around for a while. The first one was the Aerobee sounding rocket in 1947. It reached 117500m. One of the smaller Aerobee variants of the 1950s was about this size. There have been many small sounding rockets over the decades; the UK and Australia launched a lot of them.
Some intellectual property lawyer from UC Berkeley has been quoted as saying that Berkeley should have patented TCP/IP. (Can anyone find this quote). UC Berkeley didn't either design TCP/IP or do the first TCP/IP implementation. They did the fourth or fifth. (BBN, Phil Karn, Dave Mills, and 3COM all had earlier implementations.) They didn't even have the first UNIX implementation; that was 3COM's UNET. What BSD had was government funding to give their implementation away. UNET was about $3K per machine.
The trouble with "malicious URL scanners" that look for malware is that unless they're real-time, they're too late. The lifetime of bad sites is now often measured in hours.
Still, continually detecting the bad guys and beating on them does have effect. Major services have to do it, or they get pwned.
We do some tracking of major sites being exploited by phishers, There are only 29 sites on the list today, one of the shortest lists we've had in years. It's been as high as 140. The URL-shortening sites get hit daily, and kick most of the bad guys off within hours. The free-hosting sites get hit too, and most of them are good about kicking the phishers off. (Piczo and webs.com are not too good at this. t35 has gotten much better. Google has a big problem with people using documents and spreadsheets as phishing sites.)
This makes Google's browser hostile code. It should not be allowed through corporate firewalls. On the browser front, progress has been made by giving parts of the browser that run external code less privilege. Sandboxing Flash and Acrobat Reader is progress. Mozilla's dividing of add-ons into a non privileged content script and a somewhat more privileged add-on code is progress. Putting an equivalent of Back Orifice into a browser is not.
The announcement says: the technology right now is limited so that permission must be granted each time remote administration is activated. How long will that last? Could be changed silently by a forced update? What if law enforcement wants to use it? Does the remote session run through a Google server? (The protocol is apparently based on Google Talk, which does.) How else do they get two clients behind DHCP routers talking to each other? Is the connection encrypted? Is it encrypted end to end, or is the server in a position to mount a man in the middle attack? Does Google commit contractually to not accessing your machine, or is there an EULA that says they can do that whenever they want to?
If you want remote desktop access in the corporate environment, there are management tools for that. They're usually locked down tightly, since they're inherently a security risk.
Right now, we have two main models for storage - files and a flat address space. Neither is well suited to flash memory, let alone something like memristors.
There are other architectures. Burroughs machines from 1960 onward had memory addressing that worked like pathnames. Think of memory addresses as being hierarchical, like "process22/object21/array4[5]". Objects were paged in and out, but were not persistent. The IBM System/38 went further and made such objects persistent. However, such architectures are not compatible with C programming, which assumes that addresses are numbers.
Flash is usually treated as disk, even though it has an access time that's faster than the time the OS takes to grind through the file system code. Flash is much more of a random access device, but it's seldom supported as one. One possibility would be to support flash memory as a name/value tuple store, like the "non-SQL" databases.
PC-type architectures still don't have a channel architecture (like IBM mainframes since 1967), where a user program can have non-privileged access to a device without the OS being in the middle. (Channel support means a process can talk to a device directly by sending messages, messages which the device can securely associate with the sending process and its privileges. The device has to maintain security and has the information to do so.) So smart devices that do lots of little fast transactions, like a tuple store or an object store, have to be mediated through the OS. So most OSs are still reading big blocks from flash, then caching them in RAM. That's already limiting, and memristors make it more so.
The success of C and the UNIX model (which even Windows NT and its successor follow) is based on a vanilla hardware model. That's holding computing back.
Obviously they don't have a "magnetic beam" problem. They might have some electrical wiring problem. Those are easy to find.
Only once have I seen a real "magnet problem". I was trying to get a flux-gate compass to work in a mobile robot at Stanford, and was getting bogus results. So I got an ordinary needle compass, and observed that it didn't point north. I walked around with the compass plotting directions, and the center of the problem was a small building about a block away. When I went into the building, I saw "High Magnetic Field" warning signs, and found out there was a superconducting magnet in there. A big one. Even there, it was only one gauss just outside the lab.
His "feedparser" site is down, but the software is still at Google Code, and there are other maintainers.
The E-Cat is deceptively simple: hydrogen is passed over a special catalyst based on nickel in a container about a litre in size, and enough heat is produced to boil water. A demonstration in January appeared to show a several kilowatts of output from a four hundred watt input.
From the picture, which shows tubes marked "H" and "O2", this looks like a system for converting hydrogen and oxygen into water and heat with a catalyst. That's a routine technology; after all, you can burn H and O2; a catalyst just lets you do it at a lower temperature. This doesn't appear to be a closed system; for that they'd have to crack the water produced (they get steam out) back into H and O2.
How are the volumes of H and O2 going in measured? How is the heat coming out measured? There's some detail on NyTeknik but not much. It's not clear why there's 30Kg of lead in the thing. I wonder what's inside that lead. It's supposedly not producing any high-energy particles. If it did, that would be interesting. They don't seem to have radiation detectors around.
(Some years ago, when the Pons- Fleischmann cold fusion flap was underway, I went to a talk at Stanford by some physicists who were trying to reproduce the experiment. They'd started out with radiation detectors and alarms surrounding the apparatus in case it suddenly produced dangerous amounts of radiation. After a while, it was clear that nothing dramatic was going to happen. They were trying to measure neutron output, and background radiation was more than whatever the cold fusion apparatus was putting out. They finally put it in a big box of lead cubes to get rid of background neutrons, and still couldn't measure any neutrons coming out. General feeling of exasperation after weeks of work.)
All these short tests of an hour or day are suspicious. If this thing really worked, they could set it on a glass table (so observers could check for external connections) and run it for a month. The short runs hint that some consumable is being used up.
(I first posted this in 2002.)
Space travel with chemical fuels isn't feasible.
After half a century of building big rockets, we now know that they don't work very well. Half a century ago, they were use-once-and-throw-away devices, and they still are. Payloads are still tiny compared to the launch weight, even for the Shuttle. Compare the figures for jet aircraft, which can be half payload.
Reliability is still lousy, too. This is because so much weight reduction is required just to get the things off the ground that they don't have adequate safety margins. About 10-20% of satellite launches still fail, almost half a century after the first one. That number isn't improving, either; in fact, it was a little better in the 1970s. There have only been a few hundred Shuttle flights, and it's crashed once. (Update since I wrote this in 2002: twice). Commercial aircraft flights, by comparison, fail a few times per year, out of millions of flights.
Half a century in aviation took us from the Wright Brothers Flyer to the B-52. Half a century in rocketry took us from the Atlas I to the Atlas V. There's been little progress in launch vehicles since the 1960s. All the major launch systems were created decades ago. So chemical fuels just don't have the power-to-weight ratio for useful space travel. People knew this in the Orion nuclear rocket days; it's a straightforward calculation. It's unfortunate that an Orion wasn't launched once or twice, just to demonstrate that nuclear propulsion is possible.
The bullying they're concerned about mostly involves gay kids. From the paper: "Jamey Rodemeyer, a 14-year old boy from Williamsville, near Buffalo, NY, took his life after what his parents claim was yeas of bullying because of struggles with his sexuality." "Tyler Clementi, an 18-year-old Rutgers student, committed suicide by jumping off the George Washington Bridge after two classmates secretly taped him during a sexual encounter with a man and broadcast it over the web."
One of the papers cited is "LGBT and Allied Youth Responses to Cyberbullying: Policy Implications" On page 128 (the paper starts at journal page 115), after discussing how big a problem this is for gay youth, they remark "This finding somewhat contradicts a study by Smith, et al. (2008) on cyber- bullying in secondary school, which did not factor the characteristics of sexual identity and gender identity into the equation. Instead, they found that secondary school students, in general, recommended as their best coping strategies in coun- tering cyberbullying both blocking and avoiding messages, and telling someone when they were being cyberbullied." The key point here is that relatively simple avoidance strategies work for hetrosexual kids, while gay kids have much worse problems. They often can't get help from their parents. "Family rejection is often more feared than victimization or harassment. ... a significant percent of LGBT youth are forced to leave home
once their sexual or gender identity is questioned by family members, and approx-
imately 20-40% of all homeless youth are lesbian, gay, bisexual, or transgender".
So this really isn't a free speech problem at all. It's a parenting problem.
It's just that immutability is default, and you have to sprinkle "mutable" in your data structures where you actually need it.
That's a related idea to functional programming - single-assignment languages. The idea is that all (or at least most) names are "const" - they can only be assigned a value once. This works a lot like functional programming, but there's not so much nesting, and most things have names, rather than being anonymous within some nested expression.
Pure functional languages force programs into a tree form, with one output. A single assignment language forces programs into directed acyclic graph form, with multiple outputs, but no loops in the graph. (At least if you have some way to get multiple values returned from a function.)
Many newer "functional" languages are really single-assignment languages. This avoids getting buried in deeply nested parentheses, as with LISP.
OK, somebody posted their resume.
Python is similarly terse, and isn't statically typed. So this isn't about static typing. It's another functional programming fan.
Functional programming is a good fit to a certain class of algorithms. For the 65 programmers of a trading house, it makes sense.
Functional programming has its downsides. It tends to result in heavily nested code. It's hard to fan out results, so programs tend to be trees with a single result. Persistent state and I/O don't fit well with the functional model.
OCaml has some features that ought to be in more languages, like Djykstra's "guards" for choosing multiple alternatives.
Microsoft does it again. First they killed PlaysForSure, with its DRM, and now Zune,with its own incompatible DRM.
As I've pointed out before, the lifetime of DRM systems seems to be about five years. At the end of life, users tend to lose content, although sometimes there's a migration path.
Google Earth is really a standalone Windows application. (Remember those?) Google didn't develop it; it was from a company called Keyhole, which sold it as a service for years before Google bought them out. I had a Keyhole account back in 2003. NVidia had a promotional deal; it was cheaper with an NVidia card.
You could fly along a route in Keyhole, so a "helicopter view" isn't exactly new.
Look up the definition of cold shutdown. It doesn't matter if reactor containment is breached.
Not in this case. Here, "cold shutdown" has been redefined somewhat, to "below boiling if we can keep cooling water going in." Normally, in a cold shutdown, you can take the lid off the reactor, look inside, and replace fuel rods. They're a long way from that point.
More like: "the molten corium has burrowed deep enough to be cooled by groundwater and we are only reading 90 degrees at the twisted, melted reactor because the radioactive steam coming from below ground is dissipating the heat"
But not that bad, either. These reactors were built on bedrock. That placed them lower than would have been desirable for flood protection, but if they leak, they leak sideways, not down. There's been plenty of sideways leakage, but by now most of that water is being collected. There's now a cleaning plant in place to run the water through zeolites and catch the radioactive salts and solids. (Water itself doesn't become radioactive from exposure to gamma radiation; the longest lived radioactive isotope of oxygen has a half-life of 122 seconds.)
Now they have to figure out how to do the tough job - safely dismantling the radioactive mess in the melted core into small bits for disposal. That may take decades.