IIRC, when the Church of Scientology Internationallost a major lawsuit by Steve Fishman, the church executives turned the CSI into a shell, transferring virtually all capital and IP to the Religious Technology Center(which licensed "its" IP back to the CSI), theoretically leaving the plaintiff with nothing from which he could collect.
They'll have to be very careful about doing that. If they transfer the ownership after the court case starts (and they lose) it's quite possible that the court will declare the transfer illegal and let the seizure of assets go ahead anyway. This is what stops normal corporations and people from doing things just as shady when they get into trouble (which isn't to say that the more stupid among them don't try; trying isn't the same thing as succeeding...)
And I'd vastly recommend basic electronics classes in "why clock signals lie" and "why you use _one_ voltage, _one_ data format, and synchronize to _one_ clock signal throughout your system". The lessons of "why would I do this as a bulky, parallel transfer rather than a serial transfer" are also illuminated by having to run your own wires.
You do realize that by trying to sound smart you've shown your own limitations? Modern CPUs don't use a single clock signal; distributing it without crippling skew is too hard, and you'll be switching a bunch of hardware that doesn't need it (which costs power). Moreover skew problems also mean that high-performance communications are better done over serial transfers; you have the problem of RF emissions, but avoid the problems associated with having to keep all the signal paths the same length. (It's easier to put shielding on the system than it is to precisely match the distance travelled along lots of conductors.)
Of course, if you're only used to dealing with clock frequencies up to a few hundred MHz then these issues aren't a major problem. But don't think your long-held learnings are the whole story.
Can you imagine what would happen if this were to happen on a Monday morning at 9:00am, right before trading opened on Wall St, for any number of large financial companies?
While not every detail is clear to me, I can imagine that it would definitely involve you losing your job in the middle of a recession with damn little hope of getting a new one. I would not count on the corporate spyware and tripwires from not catching you do it either.
In short, being a dick is easy, but it is also very stupid. Why not use your time to do something productive instead?
Yes, we are. Large organizations (both private and public) are already sufficiently dependent on the internet even without the Cloud that if the 'net goes down, they're in deep trouble. Adding some Cloud to that doesn't change very much.
Of course, there's both down sides and up sides to this. The big downers are that it is potentially more expensive and you're not in control. The big up sides are that it relieves pressure on your own data center, and it's easier to trace service costs back to particular work items (this is vastly important in large organizations; smaller ones tend to have simpler finances).
Sure, with important data it is important to keep a backup somewhere else than the working copy. But that's been true for, well..., more decades than I've been programming. No change.
And only Windows can be infected by simply visiting a web site.
Apparently not. Unless Apple have fixed it in the last 48 hours, there's an open serious issue on Macs in relation to Java. OTOH, it's unclear how prevalent the exploits for this are in the wild; things that stay in the lab aren't stuff to panic over.
Alas, it seems that Win-targeted exploits are fairly common. But let's not pile on unwarranted FUD...
JEZUSFUCKINKRIST! This is the whole point of personal computers! The whole "computer revolution" thing of the 1970s, starting with the Altair, was to give people control over the data governing their lives.
I look around thirty years later and find DMCA, corporations with databanks stuffed with peoples' personal data, and people who think the internet is the only reason to own a computer. WTF?
Meh. I'm led to believe that this is just the pendulum swinging back towards centralized processing and data storage for a while, having swung quite hard towards distributed. But guess what? It will find some reason to swing back the other way in a few years. (The reason for the swing to come is probably not even guessable right now.)
Long term, what will prevail? I guess there's going to be a mix. Some stuff is really good when centralized, some is far better when distributed. Saying it's all got to be one or the other just makes one look shrill and stupid.
I personally think that copies which exist only in RAM should not be considered copies at all, but we would need the Supreme Court to reach that question to know for sure.
I'd hate to want to argue that since it is possible to construct a device which would hold persistent (at least on the scale of weeks) copies of media files purely in RAM, perhaps by mounting a ramdisk and preventing that memory from being paged out to disk. To me, it is the purpose of the copying that is important; if it is just a normal technical part of the process of playing that media file (assuming that the originating copy of the file has been legally acquired) then it is a clearly fair use, whereas putting it up on a website (however implemented) is a clearly unfair use (again, with some basic assumptions).
Patching exists for the simple reason that there is a rush to get products out the door, rather than take the time to ensure that they are secure
<sarcasm> At least we know that Duke Nukem Forever will be secure when it comes out. After all, the developers aren't ever going to push a product out of the door there in the hope that it will at least start earning them some cash... </sarcasm>
When it comes down to it, the only place there are strictly digital signals are in strictly local communications (with some exceptions like RS-232 and related and derived standards like RS-242, USB, SATA, that can run over longer distances) that exist primarily as point-to-point connections between individual ICs. And even then, when you actually look at what's being signalled on the line, the distinction between digital and analog gets harder and harder to make over the years.
High performance ICs will probably be doing detection of current flows even for runs from components that are in the same package (like the L1 cache). In many very real ways, digital is always just an approximation, a conceptual model; at the hardware level, everything is analog unless you're getting down to dealing with quantized things like single electrons. (We aren't in normal silicon devices. Not yet...)
High temperature superconductors:Remain a lab curiosity decades after solid state lasers, bright LEDs, and other lab curiosities made it into our homes.
They seem to be doing relatively well in the power transmission business and high-power magnets, but nobody's yet found a compound that works as a superconductor above freezing and which isn't toxic. Until that happens (no idea about how to predict that breakthrough) then industrial uses will be much more common than domestic ones.
Artificial Intelligence/Expert Systems: For decades expert systems have been able to outperformed doctors on diagnosis accuracy. So why hasn't the cost of medical care gone down like every other automatable vocation? Why don't doctors use these tools?
Generally, AI turned out to be much harder than everyone thought, stemming in part from the fact that nobody really knew what intelligence was in the first place. But there are a lot of systems that use results from AI research out there, quietly doing their jobs.
That they're not used to replace/support doctors in the US is a whole separate ball of wax.
Pebble bed fission.
You try building a new nuclear reactor, no matter how inherently safe, without getting lynched by the Green lobby.
Cars with small, efficient Diesel or rotary engines:GM and Mazda's teething pains gave these technologies a bad rap which hasn't been overcome 2 decades later (at least not in the U.S. market.)
Again, this is a problem that is specific to the US. Diesels are doing very well in Europe.
Nanotech as a buzzword. The pigment crystals in makeup and shampoo should not count as nanotechnology no matter what the marketing people think.
Thankfully, you have at least one point I can wholeheartedly agree with!
As far as nfs, good lord man you arent seriously trying to run nfs mounts over remote Internet connections?
Depends on whether it is NFSv4 (or later). Earlier versions are resource-hungry and insecure, and you'd better off with CIFS or AFS than NFSv3 or earlier. (I don't know v4 well enough to discuss its suitability for non-VPN WAN use.)
They aren't saying that 50% of *people* in the UK are sharing illegally, though.. they are saying that 50% of *traffic* is illegal, presumably counting by data volume. That 50% could be caused by a minority of people.
If the information in this recent image is correct, only 25% of net traffic is P2P. I guess the creatives are not that bothered about spam, and most web traffic is thoroughly mundane. All in all, there isn't much sunshine wherever that 50% came from...
Tata is pitching their homes at about 1.5 times their target market's annual income.
That's remarkably cheap. The long-term (over a century or more) multiple for western economies is in the 2.5 to 3 range. Sounds like a good investment for the buyers if they get the profile of inhabitants they expect. Which probably depends on transport options frankly (both public and private).
Curiously, the primary driver for the crash in the housing market is the fact that the price/income ratio had got too large. A multiple of 5 or more (and the credit availability to support that) just is unsustainable; the fall is merely a proper correction. Bit of a bummer if you bought at/near the top of the market though.
Is there any technical reason why you couldn't do the same [put it in the kernel] with a game engine?
No deep technical reason, but it's really hard to get code to the level of stability of an OS kernel, and game engines just aren't written to that quality. Would you want a game crash to require a hard reset of the computer and possibly a reflash of some of your hardware too? Keeping games (like other applications) in user-space makes a great deal of sense...
The US of A is a big place. Much bigger than say - the UK. Or Japan. Each of which are about the size of Texas and Oklahoma combined. The US of A is MUCH MUCH larger. You start running into economies of scale, since your HFC needs to run to individual neighborhood drops.
So the USA has a lot of land area. So? Why are you even contemplating running wire to every last bit of it? It's perfectly fair for people who insist on living out in the boonies to either put up with lower speeds or pony up for the costs of putting in a dedicated connection.
What you need to do is drop the Socialist ideal of treating everyone identically. Differentiate between cities and the countryside! That's what happens in "Socialist" Europe. Or be prepared for much higher prices or taxes. But, as a city-dweller, I'm not keen on subsidizing rural folks when they could just move back into range of existing services instead, and I reckon that others may feel the same...
Depends on which part of Virgin Media you're dealing with. The parts that used to be Telewest are mostly reasonably competent, but the parts that used to be NTL are pure evil. (Polling among my coworkers indicates that this is the case; I live in an ex-Telewest area, but they're mostly in ex-NTL areas.)
Re-engineering biological systems takes generations to debug. And a huge number of dud individuals during the development process. This is fine for tomato R&D, but generating a big supply of failed post-humans is going to be unpopular. Just extending the human lifespan is likely to take generations to debug. It takes a century to find out if something worked.
You should be aware that there are projects working toward the goal of being able to fully simulate a human at all levels of detail from the sub-cellular to the whole-body. If I remember right, they're expecting to take a generation or so to do this, but some pieces are currently under development. (For example, I remember reading about a project to simulate blood flow in real brains using a mix of advanced medical scanning and computational fluid dynamics.) As such models become more pervasive, it will become possible to work out more of the consequences of genetic changes ahead of time, and so the "debug cycle" will be shortened.
But (to give it its proper name) the Virtual Physiological Human project (meta-project really) is most certainly a long-term goal; if someone manages it in the next decade, I'll be very startled. Happy, sure, but startled.
Seriously, the Free Software world needs to call a timeout. Establish a core and devote every available resource into making that core bug free and secure. Then allow no change to be committed to that core without extensive peer review to prevent new bugs from getting in.
Some projects have worked that way for a long time. They still have bugs, but they tend to be of the "Oh, that probably shouldn't work that way" kind rather than the "Oh, that buffer overrun looked bad" kind. But it is equally important to keep on accepting new features; stagnation is death because developers abandon the project.
How does one even know if a particular piece of documentation, sketchy and incomplete as it will certainly be, documents what was, what currently is or what is intended to be?
I like the rule that "the code isn't done until it's got automated tests and documentation". Tests mean that new bugs are at least kept somewhat under control (even if they're not a proof of correctness, they at least catch silly errors) and docs mean that people can use it. And yes, it's often worthwhile hiring a tech writer to get the docs usable by people outside the group developing the software; it is far too easy to be blind to inexplicable quirks and odd workarounds...
Under this, you are no longer, using their own definition, "stealing" when using p2p networks. You pay their licenses.
I think that's a bit of a reach; the definition says nothing at all about making anything that you can download (on an episode-by-episode basis) legal to access. Instead, I read the clauses as explicitly excluding "on demand" services and sites like YouTube. The tax will instead be applying to the general broadcast model, which is adaptable to the internet (the technology needed is multicast, which is getting better supported than it used to) but is still primarily OTA.
An actual threat is one thing - and it's already covered in current law. So is a conspiracy to commit a crime. But saying that all XXX's ought to be killed? That's free speech.
Or incitement (i.e., attempting to induce others to commit murder). There is a long-standing principle in law that trying to get other people to commit crimes on your behalf is also a crime (with circumstances dictating whether this is conspiracy or incitement) and I'd be startled if your right to free speech totally overrides this even in the US. After all, the SCOTUS has held that there are circumstances where "free speech" is in fact law-breaking (the "FIRE in a crowded theater" example is a classic).
Britain should build power stations where the waste heat (and CO2) is used for agriculture.
I believe that there are already plans to do just this, or to use the heat for industrial processes instead, which is also a reasonable thing to do. (Save one way or save another.) Tracking down specifics is a bit trickier to do due to the amount of blog noise kicked up by the eco-nutjobs.
Slashdot Mirror
IIRC, when the Church of Scientology Internationallost a major lawsuit by Steve Fishman, the church executives turned the CSI into a shell, transferring virtually all capital and IP to the Religious Technology Center(which licensed "its" IP back to the CSI), theoretically leaving the plaintiff with nothing from which he could collect.
They'll have to be very careful about doing that. If they transfer the ownership after the court case starts (and they lose) it's quite possible that the court will declare the transfer illegal and let the seizure of assets go ahead anyway. This is what stops normal corporations and people from doing things just as shady when they get into trouble (which isn't to say that the more stupid among them don't try; trying isn't the same thing as succeeding...)
And I'd vastly recommend basic electronics classes in "why clock signals lie" and "why you use _one_ voltage, _one_ data format, and synchronize to _one_ clock signal throughout your system". The lessons of "why would I do this as a bulky, parallel transfer rather than a serial transfer" are also illuminated by having to run your own wires.
You do realize that by trying to sound smart you've shown your own limitations? Modern CPUs don't use a single clock signal; distributing it without crippling skew is too hard, and you'll be switching a bunch of hardware that doesn't need it (which costs power). Moreover skew problems also mean that high-performance communications are better done over serial transfers; you have the problem of RF emissions, but avoid the problems associated with having to keep all the signal paths the same length. (It's easier to put shielding on the system than it is to precisely match the distance travelled along lots of conductors.)
Of course, if you're only used to dealing with clock frequencies up to a few hundred MHz then these issues aren't a major problem. But don't think your long-held learnings are the whole story.
Can you imagine what would happen if this were to happen on a Monday morning at 9:00am, right before trading opened on Wall St, for any number of large financial companies?
While not every detail is clear to me, I can imagine that it would definitely involve you losing your job in the middle of a recession with damn little hope of getting a new one. I would not count on the corporate spyware and tripwires from not catching you do it either.
In short, being a dick is easy, but it is also very stupid. Why not use your time to do something productive instead?
Over the internet? Are we really THAT insane?
Yes, we are. Large organizations (both private and public) are already sufficiently dependent on the internet even without the Cloud that if the 'net goes down, they're in deep trouble. Adding some Cloud to that doesn't change very much.
Of course, there's both down sides and up sides to this. The big downers are that it is potentially more expensive and you're not in control. The big up sides are that it relieves pressure on your own data center, and it's easier to trace service costs back to particular work items (this is vastly important in large organizations; smaller ones tend to have simpler finances).
Sure, with important data it is important to keep a backup somewhere else than the working copy. But that's been true for, well..., more decades than I've been programming. No change.
And only Windows can be infected by simply visiting a web site.
Apparently not. Unless Apple have fixed it in the last 48 hours, there's an open serious issue on Macs in relation to Java. OTOH, it's unclear how prevalent the exploits for this are in the wild; things that stay in the lab aren't stuff to panic over.
Alas, it seems that Win-targeted exploits are fairly common. But let's not pile on unwarranted FUD...
JEZUSFUCKINKRIST! This is the whole point of personal computers! The whole "computer revolution" thing of the 1970s, starting with the Altair, was to give people control over the data governing their lives.
I look around thirty years later and find DMCA, corporations with databanks stuffed with peoples' personal data, and people who think the internet is the only reason to own a computer. WTF?
Meh. I'm led to believe that this is just the pendulum swinging back towards centralized processing and data storage for a while, having swung quite hard towards distributed. But guess what? It will find some reason to swing back the other way in a few years. (The reason for the swing to come is probably not even guessable right now.)
Long term, what will prevail? I guess there's going to be a mix. Some stuff is really good when centralized, some is far better when distributed. Saying it's all got to be one or the other just makes one look shrill and stupid.
I don't want Big Brother in my car for the same reasons I don't want big Brother in my computer.
If it's not logging or reporting home, it's hardly Big Brother.
I personally think that copies which exist only in RAM should not be considered copies at all, but we would need the Supreme Court to reach that question to know for sure.
I'd hate to want to argue that since it is possible to construct a device which would hold persistent (at least on the scale of weeks) copies of media files purely in RAM, perhaps by mounting a ramdisk and preventing that memory from being paged out to disk. To me, it is the purpose of the copying that is important; if it is just a normal technical part of the process of playing that media file (assuming that the originating copy of the file has been legally acquired) then it is a clearly fair use, whereas putting it up on a website (however implemented) is a clearly unfair use (again, with some basic assumptions).
<sarcasm>
At least we know that Duke Nukem Forever will be secure when it comes out. After all, the developers aren't ever going to push a product out of the door there in the hope that it will at least start earning them some cash...
</sarcasm>
When it comes down to it, the only place there are strictly digital signals are in strictly local communications (with some exceptions like RS-232 and related and derived standards like RS-242, USB, SATA, that can run over longer distances) that exist primarily as point-to-point connections between individual ICs. And even then, when you actually look at what's being signalled on the line, the distinction between digital and analog gets harder and harder to make over the years.
High performance ICs will probably be doing detection of current flows even for runs from components that are in the same package (like the L1 cache). In many very real ways, digital is always just an approximation, a conceptual model; at the hardware level, everything is analog unless you're getting down to dealing with quantized things like single electrons. (We aren't in normal silicon devices. Not yet...)
High temperature superconductors:Remain a lab curiosity decades after solid state lasers, bright LEDs, and other lab curiosities made it into our homes.
They seem to be doing relatively well in the power transmission business and high-power magnets, but nobody's yet found a compound that works as a superconductor above freezing and which isn't toxic. Until that happens (no idea about how to predict that breakthrough) then industrial uses will be much more common than domestic ones.
Artificial Intelligence/Expert Systems: For decades expert systems have been able to outperformed doctors on diagnosis accuracy. So why hasn't the cost of medical care gone down like every other automatable vocation? Why don't doctors use these tools?
Generally, AI turned out to be much harder than everyone thought, stemming in part from the fact that nobody really knew what intelligence was in the first place. But there are a lot of systems that use results from AI research out there, quietly doing their jobs.
That they're not used to replace/support doctors in the US is a whole separate ball of wax.
Pebble bed fission.
You try building a new nuclear reactor, no matter how inherently safe, without getting lynched by the Green lobby.
Cars with small, efficient Diesel or rotary engines:GM and Mazda's teething pains gave these technologies a bad rap which hasn't been overcome 2 decades later (at least not in the U.S. market.)
Again, this is a problem that is specific to the US. Diesels are doing very well in Europe.
Nanotech as a buzzword. The pigment crystals in makeup and shampoo should not count as nanotechnology no matter what the marketing people think.
Thankfully, you have at least one point I can wholeheartedly agree with!
You do know that crack, being insoluble, can't be snorted?
I won't tell them they've been snorting (and paying top prices for) icing sugar cut with talc, baking powder and dandruff if you won't...
As far as nfs, good lord man you arent seriously trying to run nfs mounts over remote Internet connections?
Depends on whether it is NFSv4 (or later). Earlier versions are resource-hungry and insecure, and you'd better off with CIFS or AFS than NFSv3 or earlier. (I don't know v4 well enough to discuss its suitability for non-VPN WAN use.)
They aren't saying that 50% of *people* in the UK are sharing illegally, though.. they are saying that 50% of *traffic* is illegal, presumably counting by data volume. That 50% could be caused by a minority of people.
If the information in this recent image is correct, only 25% of net traffic is P2P. I guess the creatives are not that bothered about spam, and most web traffic is thoroughly mundane. All in all, there isn't much sunshine wherever that 50% came from...
Tata is pitching their homes at about 1.5 times their target market's annual income.
That's remarkably cheap. The long-term (over a century or more) multiple for western economies is in the 2.5 to 3 range. Sounds like a good investment for the buyers if they get the profile of inhabitants they expect. Which probably depends on transport options frankly (both public and private).
Curiously, the primary driver for the crash in the housing market is the fact that the price/income ratio had got too large. A multiple of 5 or more (and the credit availability to support that) just is unsustainable; the fall is merely a proper correction. Bit of a bummer if you bought at/near the top of the market though.
Cool! I just installed [emacs] on my machine. Do you know where I can find a decent text editor for it?
It comes with one built in.
M-x vi-mode
Is there any technical reason why you couldn't do the same [put it in the kernel] with a game engine?
No deep technical reason, but it's really hard to get code to the level of stability of an OS kernel, and game engines just aren't written to that quality. Would you want a game crash to require a hard reset of the computer and possibly a reflash of some of your hardware too? Keeping games (like other applications) in user-space makes a great deal of sense...
The US of A is a big place. Much bigger than say - the UK. Or Japan. Each of which are about the size of Texas and Oklahoma combined. The US of A is MUCH MUCH larger. You start running into economies of scale, since your HFC needs to run to individual neighborhood drops.
So the USA has a lot of land area. So? Why are you even contemplating running wire to every last bit of it? It's perfectly fair for people who insist on living out in the boonies to either put up with lower speeds or pony up for the costs of putting in a dedicated connection.
What you need to do is drop the Socialist ideal of treating everyone identically. Differentiate between cities and the countryside! That's what happens in "Socialist" Europe. Or be prepared for much higher prices or taxes. But, as a city-dweller, I'm not keen on subsidizing rural folks when they could just move back into range of existing services instead, and I reckon that others may feel the same...
You don't know much about Virgin Media, do you?
Depends on which part of Virgin Media you're dealing with. The parts that used to be Telewest are mostly reasonably competent, but the parts that used to be NTL are pure evil. (Polling among my coworkers indicates that this is the case; I live in an ex-Telewest area, but they're mostly in ex-NTL areas.)
Re-engineering biological systems takes generations to debug. And a huge number of dud individuals during the development process. This is fine for tomato R&D, but generating a big supply of failed post-humans is going to be unpopular. Just extending the human lifespan is likely to take generations to debug. It takes a century to find out if something worked.
You should be aware that there are projects working toward the goal of being able to fully simulate a human at all levels of detail from the sub-cellular to the whole-body. If I remember right, they're expecting to take a generation or so to do this, but some pieces are currently under development. (For example, I remember reading about a project to simulate blood flow in real brains using a mix of advanced medical scanning and computational fluid dynamics.) As such models become more pervasive, it will become possible to work out more of the consequences of genetic changes ahead of time, and so the "debug cycle" will be shortened.
But (to give it its proper name) the Virtual Physiological Human project (meta-project really) is most certainly a long-term goal; if someone manages it in the next decade, I'll be very startled. Happy, sure, but startled.
Seriously, the Free Software world needs to call a timeout. Establish a core and devote every available resource into making that core bug free and secure. Then allow no change to be committed to that core without extensive peer review to prevent new bugs from getting in.
Some projects have worked that way for a long time. They still have bugs, but they tend to be of the "Oh, that probably shouldn't work that way" kind rather than the "Oh, that buffer overrun looked bad" kind. But it is equally important to keep on accepting new features; stagnation is death because developers abandon the project.
How does one even know if a particular piece of documentation, sketchy and incomplete as it will certainly be, documents what was, what currently is or what is intended to be?
I like the rule that "the code isn't done until it's got automated tests and documentation". Tests mean that new bugs are at least kept somewhat under control (even if they're not a proof of correctness, they at least catch silly errors) and docs mean that people can use it. And yes, it's often worthwhile hiring a tech writer to get the docs usable by people outside the group developing the software; it is far too easy to be blind to inexplicable quirks and odd workarounds...
Under this, you are no longer, using their own definition, "stealing" when using p2p networks. You pay their licenses.
I think that's a bit of a reach; the definition says nothing at all about making anything that you can download (on an episode-by-episode basis) legal to access. Instead, I read the clauses as explicitly excluding "on demand" services and sites like YouTube. The tax will instead be applying to the general broadcast model, which is adaptable to the internet (the technology needed is multicast, which is getting better supported than it used to) but is still primarily OTA.
An actual threat is one thing - and it's already covered in current law. So is a conspiracy to commit a crime. But saying that all XXX's ought to be killed? That's free speech.
Or incitement (i.e., attempting to induce others to commit murder). There is a long-standing principle in law that trying to get other people to commit crimes on your behalf is also a crime (with circumstances dictating whether this is conspiracy or incitement) and I'd be startled if your right to free speech totally overrides this even in the US. After all, the SCOTUS has held that there are circumstances where "free speech" is in fact law-breaking (the "FIRE in a crowded theater" example is a classic).
Britain should build power stations where the waste heat (and CO2) is used for agriculture.
I believe that there are already plans to do just this, or to use the heat for industrial processes instead, which is also a reasonable thing to do. (Save one way or save another.) Tracking down specifics is a bit trickier to do due to the amount of blog noise kicked up by the eco-nutjobs.
I'd hate to wolfram stuff.
He should change the name of the search engine to Tungsten.
I'll get my hat.