It cracks me up to keep seeing states jumping through hoops and giving away all sorts of tax revenues for these big companies to set up shop. Then, later on, the company reveals that only about 30 jobs are going to be created in actuality, and the state has lost more than if they had just let the deal pass them by.
Does this include the verbal open offers for officials who helped secure the deal?
Me too! This is totally, positively, awesome. I looked through the code and it should port super easy to any system that can provide a libusb-like host interface to the USB bus. Which isn't exactly tough. First I'm going to use it to drive a display using an Olimex LPC-E2468 running plain uCLinux just to check it out; then I plan to port it to my own (MIT license) networking RTOS. This is exactly what I have been looking for!
Second, Debian (as a stock install, I don't include remastered lightweight Knoppix variants in that category) does not have a significant presence in the embedded device market.
Perhaps they're trying to change this and find glibc an obstacle.
I'm assuming that not even a governmental department can be stupid enough not to have copies of the backups in a fire safe, off-site location.
This is no guarantee against datajacking. No one keeps their backups forever. If I know the retention policy is one year, then after I break in I make sure to install something, like an encrypting driver, that makes tapes and disks look fine for backup/restore - until I remove the key. I then wait a year until all backups currently retained are polluted. Then I pull the plug and make my demand. I can even do it by trigger, so when the first polluted backup is the oldest in the set the key is automatically wiped, immediately putting every bit they own on wrote from any of the infected servers up for ransom. This way, all the network traces, addresses, etc, available to track me down would also be a year old.
Just a couple of months ago, there was a huge to-do about NAT and IPv6. "IPv6 is a world without NAT". The hell it is. My internal routers don't get publicly routable IP addresses, even if I have to NAT back to IPv4.
I agree with the sentiment - however, it's one of policy, not mechanism. NAT is a pretty poor substitute for a router that implements policy (known as a firewall). NAT has literally an all-or-nothing granularity. For instance, I might want to specify that an internal host can enable BitTorrent via UPnP, but under no circumstances can CIFS be allowed through - in either direction. An internal host sending a CIFS solicitation out does not mean a pinhole should be opened and some set of hosts (depending on cone of restriction) free to respond. NAT is just not a practical policy tool. It's an address space recovery tool. Reverse NAT, however, has some redeeming qualities for load balancing and failover - I'm not versed well enough in IPv6 to understand how they'd be implemented without NAT. (Anycast addressing, I suppose.)
But you can implement NAT in IPv6 just as much as in IPv4 if you wish. A router could appear to have a single interface ID and translate to/from that. It's largely unnecessary though since instead of a handful of IPv4 addresses you have an entire 64-bit space to yourself (and maybe even the SLN prefix, not sure about that).
IPv6 really is a major cleanup and simplification from IPv4. I'm slightly disconcerted by the increased dependency on DNS however.
0. Evil Bolivian liberals start talking about using the proceeds from sale of lithium for things like national defense, highways, electricity, water plants, schools, research facilities, health care, a functional judicial system - all this first-world stuff they could only dream of affording previously 1. Coup 2. Generals clean out subversives who think Bolivians should own their own natural resources, and make country safe for U.S. and European mining co's 3. Generals sell off complete and exclusive rights for pennies on the dollar - no taxation or local businesses involved; Generals get rewarded with nice personal kickbacks 4. Generals provide local labor for cheap. Very cheap. After all, they have a virtually infinite supply of desperate people willing to work for subsistence wages 5. After 10-20 years as the locals revolt because of the total sell-out, generals escape to a first-world life in luxury 6. As the locals refuse to accept the previous BS deal they kick out foreign mining co's and nationalize the resources 7. U.S. decries evil commies and does its best to destabilize said evil commie government, by interfering with elections, supporting "freedom fighters" (read: insurgents and terrorists), and generally attempt to turn back the clock. The pretext is demanding "free elections", which of course can be rigged to practically restore the previous order 8. After a generation everyone eventually gets tired of conflict, forget what they were fighting over in the first place, and things are allowed to return to some semblance of where they should have been at point 0. Only with a lot of bad history.
Without use of Kazakhstan, Russia has only a narrow strip of land that stretches far enough south to be worth launching from - and landing at. And this is not a flat desert wasteland. The reason for the rockets is to allow for a controlled landing. Parachutes are more suited for an ocean or desert landing where a few miles of accuracy doesn't make much difference. Presumably they figured that the weight of the landing system is outweighed by the benefit of launching (and landing) at a more southern latitude. Ocean landings aren't exactly free, either.
I hacked Minix a new memory manager in a System Programming at University class back in 1996. I'm quite literally apathetic with incredulity that the EU are funding further development.
This was my initial reaction too... As in what, that thing is still around? Does it still use 8086 segment registers for memory management? I figured someone must have done an excellent sales job to fund work on Minix!!!
But then, it's about security research and Minix is presumably only used as the vehicle. This makes more sense. Today a hodge podge of security models are used, all inadequate in one way or another.
Say you want to grant libpcap the ability to put an interface into promiscuous mode, but limit use of libpcap to ethereal, then limit the users who can run ethereal, and maybe even allow them to do different things with it. In addition, you want to authenticate executable and library signatures. And you want all this to be done without significant implementation in the libraries and executables themselves - you want them to implement functionality, not security. The security part is orthogonal and needs to intersect all functionality. A microkernel like Minix seems like a good vehicle for this, because so much system functionality is implemented in userland. VAX/VMS was pretty good about finely granular privileges, but it didn't really have a generalized model that was extensible or even all that suited for non-monolithic kernels. In fact, part of what keeps us with monolithic kernels is the difficulty of securing microkernel based systems. (Not really including single server ones, like OS X.) Of course, performance is another problem but I think that can be overcome. (Mach just screwed it up by making it too high level.)
So, weird as it may seem at first glance, it does make some sense. I think. I would welcome a research system that demonstrates a generalized - and practical - model for capability management.
Agreed. Time to abandon all the legacy crap. Apple did the same in OS X, where for a while they bundled OS 9 in an emulated environment. (Aka the Blue Box.)
can you run *nix on your hardware, then get hardware accelerated Direct3D and OpenGL in Windows running in a VM?
Yes, if the VM vendor offers an D3D/OGL graphics driver for Windows. (VMware does for Windows on OS X, for instance.) There is some performance loss in the translation though.
I would possibly like to see such a public network run as a wholesale service whereby the service providers buy capacity and resell it with their own packages.
The way I see it it has nothing to do with socialism or private vs government. It has to do with a division of who does what. Clearly no free market exists without a government (the fabled bazaar is very much a government creation), and corporations are communist: they're command economies strictly planned and executed from top, ultimately controlled by a small circle (the politburo/board of directors). And like any communist structure they give lip service to things like integrity and values while planning to stab each other in the back. It's not surprising that organizations like this can't meet deliverables while they enrich the most important people in the world: their senior management. (And even if they're too humble in public to express this sentiment, that's how they view themselves.)
The problem with broadband today is that the infrastructure model is that of railroads. A RR could hold a municipality hostage and make preposterous demands in return for running a line to it. Even though running the line without any incentive whatsoever from the local government would still be profitable and a very good business idea, milking the municipality was even better! Make a big public outcry about how expensive and unprofitable it is and how it needs to be supported with tax incentives and exclusive deals. And we all saw how well the RR's handled changes in infrastructure needs. Nationalizing didn't really make much difference, because the fundamental problem is in the infrastructure model itself: competing behemoths that own and manage infrastructure as property.
What we need is the interstate model. Yeah, it's not profitable. Heck, we don't even try to monetize it. Instead we recognize that it provides value greatly in excess of the tax payer cost of operating it. It's open to everyone for whatever they choose to use it for. If you want to run a trucking company, by all means, all you need is a truck! You don't have to start by building roads, or leasing rights to whatever stretch you want to run. Just do it, who knows, maybe you can grow it to compete with UPS? Yeah, there may be weight stations, gasoline taxes, and fees, but that's really not a significant issue, and it's the same for everyone. No old-boys network, no special deals, no exclusive contracts. In providing the interstate system we provide an important tool to reduce the barrier of entry for transportation. We help create a market for transportation services.
What if UPS and Fedex owned the interstate system? Would anyone in their right mind think it would be a good idea to sell it in pieces to the highest bidder? Yet, that is exactly what we've done with our communications infrastructure. Of course it's dysfunctional. Trying to regulate free access back into it really doesn't work either - because of the nature of authoritarian command economies and how the people at their top think it simply isn't in their nature to accept.
We've done the same for airports - these are paid for by taxpayers, and even though there are some usage fees, it's pretty trivial stuff. Want to start an airline? By all means; all you need is terminal space and an aircraft. Airport operations and air traffic control is provided by the tax payer.
So we should create a federal communications authority, that simply goes ahead and lays the fiber. It should contract not with telcos but with the people who actually dig the trenches and splice the ends. Then they sell access at some nominal fee, not necessarily to make a profit or even break even but to avoid the tragedy of the commons. The resulting economic growth spurred by innovation in technology and new previously impossible business models then produces a tax base which indirectly pays for the infrastructure. This is really the ONLY way to get it done. Unfortunately our
With this technique you can cross-infect VMs and rootkit the hypervisor. That's a much bigger problem than simply rolling back a VM to a clean snapshot.
Is the industry good at picking stuff that becomes popular and sells? Or does it become popular because the industry tells us to like it and provides no alternative?
Music by its very nature is "sticky". If you hear something long enough you will become familiar with it and start liking it. (If not, then it's not musical.) The more similar it is to previously familiar music, the easier it sticks. What the major labels look for is music that is sufficiently similar to current mainstream to be cheap to make stick (say play five times over a week on radio, as opposed to 30 times over two months), while having at least some trivial differentiation and non-music appeal. Be it sexiness, dance skills, rebellion, or some other unrelated appeal.
Music doesn't spread by "word of mouth". It spreads by your friends playing it over and over while you're in their presence. It's a push process, not pull.
It's not surprising that teenagers and kids in their 20s download for free - because they can't afford to buy everything they've developed familiarity with. They lack the means to carry to the entertainment industry on their shoulders. And not having access to popular culture pretty much makes you a social outsider - not good for the development of teenagers, or the life of 20-somethings when so much hinges on social drama. Those of us who are older certainly pay for our music, but we can also afford to.
The popularity of a cultural work is largely a result not of any inherent qualities of the work itself, but of of the activities of the audience.... This is a power law distribution with its long tail.
This applies to all aspects of human culture, not just music taste. This is why totalitarian thought control works: if you consistently eliminate or suppress the impopular 1-2% outliers on the curve the curve will narrow. But it retains it shape, so there is always an impopular fringe the vast majority might only miss "in principle". Principle being the classic liberal freedoms: of expression, equality before the law, absence of tyranny, and such abstract concepts. At least until integrity and principle itself is suppressed.
The record labels and the entertainment industry accomplish exactly the same result - whether wittingly or by coincidence. They target the tastes near the center of the curve and won't touch the outliers. Hence they produce a cultural narrowing; not many people would say they miss the outliers "because no one they know likes that crap anyway", other than perhaps in principle. The principle being that cultural plurality is inherently beneficial. The entertainment industry is good at producing choice, but with few alternatives. It's the same reason a walk down the supermarket bread isle reveals 30 kinds of breads, none with rye listed as a first ingredient, and only one or two without syrup, honey, or milk solids. Plurality of choice in the absence of alternative.
Powell's only sells books though. Only a tiny fraction of what I spend through Amazon is actually on books. I too would like a viable-one stop alternative.
When it comes to measurements, I'm much more interested in how long it takes from when you open the lid to when you can check you email or load a web page. So wake time, plus wifi auto connect time, plus DHCP config time, plus browser/mail connection and load/rendering time. Also, I'd be interested to know how long it will not only run on a charge, but also how long it will sleep on one, and whether I can swap batteries while it's sleeping and have it stay asleep. Assuming the battery is replaceable of course.
Most of the "performance" above is not hardware related at all, but software related, like how the wifi drivers and network protocols like DHCP are implemented, and how well the stack is vertically integrated. They're somewhat dependent on performance and capabilities provided by the hardware though.
For a laptop I think this is much, much, much more important than how well it plays some game.
DOCSIS is TDMA, which just doesn't play nice with TCP congestion control. This is why once the time slots become more intermittent as utilization goes up TCP performance tanks.
It doesn't really matter how much CO2 it emits. You grow something, then eat it, and then reemit it. The carbon then goes back to being absorbed by something growing - it's a closed system powered by the sun.
The increase in atmospheric carbon comes from us digging coal out of hillsides and pumping oil from underground, then burning it - adding it to the cycle. But it has nowhere to go in the short term, as in a few hundred or thousand of years, so results in increased atmospheric concentration.
Things that grow, live, eat, die, don't affect the net balance of atmospheric carbon. It doesn't matter how much you breathe, fart, your cows fart, or whether you brew beer. 1 lb of carbon in plant matter in can't result in more than 1 lb of carbon out or vice versa. It also makes no difference if it's CO2 or methane - the net result is for all practical purposes the same. (Methane is what, 15x more potent, but CO2 lingers about 18x longer; so over their lifetime it's a pound-for-pound toss-up.)
Likewise, if you ask a question of the form, "ARE zebras mammals?", in order to retrieve the answer, an algorithm that precisely represents an implementation of "Is this proposed assertion true" in the context of the specific data encoding must be run.
But for a SEARCH ENGINE, what you really want is not an answer to the question, but a list of results that are relevant to the reasoning involved in determining the answer.
The problem here is that the answer can vary depending on context, even a simple one like the one above. For instance, to everyday people zebras are Savannah Horsies, and horsies are mammals, so yes they _suppose_ this is correct. If you ask them, however, they might wonder if it's a trick question and may feel unsure. Biologically, zebras are mammals by taxonomy. Philosophically... what "is a mammal" (or taxonomy as such for that matter)? Is it a genetic pattern? Order of descent? Kinship through descent? Physical attributes? Is a non-mammal zebra, as a thought experiment, impossible? Is the mammal-ness merely a heuristic decision agreed to by a scientific quorum and if so could it be challenged?
What you might want from a search engine is anything related to answering the question - because presumably you're looking to deepen your understanding of something.
Because Steve Bourne was doing this work back in 1975! About that time people had only just got beyond programming by biting holes in paper-tape with their teeth!
There were certainly vastly more powerful command interpreters around. In fact, sh is a pretty crappy interface and was even then; its strength is in the ability to chain stdin/stdout and it simplified and cleaned up the system considerably by performing globbing. For a pure command line user interface, TOPS-20 was the gold standard in 1975. Sh's strength wasn't in its user friendliness, but in its scripting ability. Bash uses readline, but readline still lacks syntactic awareness.
Because the swiss have some sort of problem with a foreigner just waltzing into their country and taking a job when they've already got swiss that can do it.
As a scientist with a PhD and in your early 30s (i.e. with about 10 years' experience) you would: 1. Fly to Switzerland and stay up to 90 days without a visa if you're American/Canadian 2. Find employment in your field. 3. Have the employer obtain a specialist visa. 4. Work ten months. 5. Get an unlimited "C" work permit/visa.
With a "C" visa you're no longer tied to that particular employer but can move freely on the labor market.
Much easier than if you're an Indian PhD who wants to work in the U.S. - or a Swiss one for that matter. H1Bs are far more restrictive, and there is nothing like the Swiss "C" visa for when you're somewhere between H1B and permanent residency (at 48months IIRC). You also won't have to jump through silly administrative hoops, like go back to your home country to apply for the visa. And the Swiss will be nice, polite, and actually helpful at the border and immigration offices. As opposed to the snarling, incompetent, rude, bottom-of-the-barrel idiots you encounter here. If there's something wrong with your application or paperwork they will helpfully suggest how to correct it. Again, very different from U.S. government standards.
The fundamental flaw in the government's plan is this: in order for a wireless base station or home router to log, it has to log somewhere else. That means a computer with a static IP that must always be on.
Not really. My, by now rather antiquated, Netgear FVS318 firewall/NAT router will send a daily log by email. It can be set to log just about everything, including every connection made. If it has lots of logging to do it will send email more frequently. I sort this out using procmail and run it through a filter.
I propose if this bill passes people buy devices like this. Then set it to fully verbose logging and point it at: postmaster@gop.org. Problem solved - they get all the logs they could conceivably want and can safekeep their precious security hoard for themselves.
Nor does the government have the right to limit my consumption. The only things that should limit my consumption are my available resources and self control.
I don't think anyone cares about your consumption. We only care that you don't poison the village well - because if you do despite us repeatedly warning you not to, we'll string you up by your balls, cut you to little pieces, and use you for dog food.
Drive your SUV, but feed the tailpipe into the cabin and breathe your own toxic filth. Nobody cares if you do - just don't blow your crap into the same air the rest of us breathe.
Slashdot Mirror
It cracks me up to keep seeing states jumping through hoops and giving away all sorts of tax revenues for these big companies to set up shop. Then, later on, the company reveals that only about 30 jobs are going to be created in actuality, and the state has lost more than if they had just let the deal pass them by.
Does this include the verbal open offers for officials who helped secure the deal?
Me too! This is totally, positively, awesome. I looked through the code and it should port super easy to any system that can provide a libusb-like host interface to the USB bus. Which isn't exactly tough. First I'm going to use it to drive a display using an Olimex LPC-E2468 running plain uCLinux just to check it out; then I plan to port it to my own (MIT license) networking RTOS. This is exactly what I have been looking for!
Second, Debian (as a stock install, I don't include remastered lightweight Knoppix variants in that category) does
not have a significant presence in the embedded device market.
Perhaps they're trying to change this and find glibc an obstacle.
I'm assuming that not even a governmental department can be stupid enough not to have copies of the backups in a fire safe, off-site location.
This is no guarantee against datajacking. No one keeps their backups forever. If I know the retention policy is one year, then after I break in I make sure to install something, like an encrypting driver, that makes tapes and disks look fine for backup/restore - until I remove the key. I then wait a year until all backups currently retained are polluted. Then I pull the plug and make my demand. I can even do it by trigger, so when the first polluted backup is the oldest in the set the key is automatically wiped, immediately putting every bit they own on wrote from any of the infected servers up for ransom. This way, all the network traces, addresses, etc, available to track me down would also be a year old.
...is that the estate of Gene Roddenberry hasn't gotten involved yet.
Just a couple of months ago, there was a huge to-do about NAT and IPv6. "IPv6 is a world without NAT". The hell it is. My internal routers don't get publicly routable IP addresses, even if I have to NAT back to IPv4.
I agree with the sentiment - however, it's one of policy, not mechanism. NAT is a pretty poor substitute for a router that implements policy (known as a firewall). NAT has literally an all-or-nothing granularity. For instance, I might want to specify that an internal host can enable BitTorrent via UPnP, but under no circumstances can CIFS be allowed through - in either direction. An internal host sending a CIFS solicitation out does not mean a pinhole should be opened and some set of hosts (depending on cone of restriction) free to respond. NAT is just not a practical policy tool. It's an address space recovery tool. Reverse NAT, however, has some redeeming qualities for load balancing and failover - I'm not versed well enough in IPv6 to understand how they'd be implemented without NAT. (Anycast addressing, I suppose.)
But you can implement NAT in IPv6 just as much as in IPv4 if you wish. A router could appear to have a single interface ID and translate to/from that. It's largely unnecessary though since instead of a handful of IPv4 addresses you have an entire 64-bit space to yourself (and maybe even the SLN prefix, not sure about that).
IPv6 really is a major cleanup and simplification from IPv4. I'm slightly disconcerted by the increased dependency on DNS however.
0. Evil Bolivian liberals start talking about using the proceeds from sale of lithium for things like national defense, highways, electricity, water plants, schools, research facilities, health care, a functional judicial system - all this first-world stuff they could only dream of affording previously
1. Coup
2. Generals clean out subversives who think Bolivians should own their own natural resources, and make country safe for U.S. and European mining co's
3. Generals sell off complete and exclusive rights for pennies on the dollar - no taxation or local businesses involved; Generals get rewarded with nice personal kickbacks
4. Generals provide local labor for cheap. Very cheap. After all, they have a virtually infinite supply of desperate people willing to work for subsistence wages
5. After 10-20 years as the locals revolt because of the total sell-out, generals escape to a first-world life in luxury
6. As the locals refuse to accept the previous BS deal they kick out foreign mining co's and nationalize the resources
7. U.S. decries evil commies and does its best to destabilize said evil commie government, by interfering with elections, supporting "freedom fighters" (read: insurgents and terrorists), and generally attempt to turn back the clock. The pretext is demanding "free elections", which of course can be rigged to practically restore the previous order
8. After a generation everyone eventually gets tired of conflict, forget what they were fighting over in the first place, and things are allowed to return to some semblance of where they should have been at point 0. Only with a lot of bad history.
Been there, done that. Got the t-shirt.
Without use of Kazakhstan, Russia has only a narrow strip of land that stretches far enough south to be worth launching from - and landing at. And this is not a flat desert wasteland. The reason for the rockets is to allow for a controlled landing. Parachutes are more suited for an ocean or desert landing where a few miles of accuracy doesn't make much difference. Presumably they figured that the weight of the landing system is outweighed by the benefit of launching (and landing) at a more southern latitude. Ocean landings aren't exactly free, either.
I hacked Minix a new memory manager in a System Programming at University class back in 1996. I'm quite literally apathetic with incredulity that the EU are funding further development.
This was my initial reaction too... As in what, that thing is still around? Does it still use 8086 segment registers for memory management? I figured someone must have done an excellent sales job to fund work on Minix!!!
But then, it's about security research and Minix is presumably only used as the vehicle. This makes more sense. Today a hodge podge of security models are used, all inadequate in one way or another.
Say you want to grant libpcap the ability to put an interface into promiscuous mode, but limit use of libpcap to ethereal, then limit the users who can run ethereal, and maybe even allow them to do different things with it. In addition, you want to authenticate executable and library signatures. And you want all this to be done without significant implementation in the libraries and executables themselves - you want them to implement functionality, not security. The security part is orthogonal and needs to intersect all functionality. A microkernel like Minix seems like a good vehicle for this, because so much system functionality is implemented in userland. VAX/VMS was pretty good about finely granular privileges, but it didn't really have a generalized model that was extensible or even all that suited for non-monolithic kernels. In fact, part of what keeps us with monolithic kernels is the difficulty of securing microkernel based systems. (Not really including single server ones, like OS X.) Of course, performance is another problem but I think that can be overcome. (Mach just screwed it up by making it too high level.)
So, weird as it may seem at first glance, it does make some sense. I think. I would welcome a research system that demonstrates a generalized - and practical - model for capability management.
Good move by MS on this. If they do it well.
Agreed. Time to abandon all the legacy crap. Apple did the same in OS X, where for a while they bundled OS 9 in an emulated environment. (Aka the Blue Box.)
can you run *nix on your hardware, then get hardware accelerated Direct3D and OpenGL in Windows running in a VM?
Yes, if the VM vendor offers an D3D/OGL graphics driver for Windows. (VMware does for Windows on OS X, for instance.) There is some performance loss in the translation though.
I would possibly like to see such a public network run as a wholesale service whereby the service providers buy capacity and resell it with their own packages.
The way I see it it has nothing to do with socialism or private vs government. It has to do with a division of who does what. Clearly no free market exists without a government (the fabled bazaar is very much a government creation), and corporations are communist: they're command economies strictly planned and executed from top, ultimately controlled by a small circle (the politburo/board of directors). And like any communist structure they give lip service to things like integrity and values while planning to stab each other in the back. It's not surprising that organizations like this can't meet deliverables while they enrich the most important people in the world: their senior management. (And even if they're too humble in public to express this sentiment, that's how they view themselves.)
The problem with broadband today is that the infrastructure model is that of railroads. A RR could hold a municipality hostage and make preposterous demands in return for running a line to it. Even though running the line without any incentive whatsoever from the local government would still be profitable and a very good business idea, milking the municipality was even better! Make a big public outcry about how expensive and unprofitable it is and how it needs to be supported with tax incentives and exclusive deals. And we all saw how well the RR's handled changes in infrastructure needs. Nationalizing didn't really make much difference, because the fundamental problem is in the infrastructure model itself: competing behemoths that own and manage infrastructure as property.
What we need is the interstate model. Yeah, it's not profitable. Heck, we don't even try to monetize it. Instead we recognize that it provides value greatly in excess of the tax payer cost of operating it. It's open to everyone for whatever they choose to use it for. If you want to run a trucking company, by all means, all you need is a truck! You don't have to start by building roads, or leasing rights to whatever stretch you want to run. Just do it, who knows, maybe you can grow it to compete with UPS? Yeah, there may be weight stations, gasoline taxes, and fees, but that's really not a significant issue, and it's the same for everyone. No old-boys network, no special deals, no exclusive contracts. In providing the interstate system we provide an important tool to reduce the barrier of entry for transportation. We help create a market for transportation services.
What if UPS and Fedex owned the interstate system? Would anyone in their right mind think it would be a good idea to sell it in pieces to the highest bidder? Yet, that is exactly what we've done with our communications infrastructure. Of course it's dysfunctional. Trying to regulate free access back into it really doesn't work either - because of the nature of authoritarian command economies and how the people at their top think it simply isn't in their nature to accept.
We've done the same for airports - these are paid for by taxpayers, and even though there are some usage fees, it's pretty trivial stuff. Want to start an airline? By all means; all you need is terminal space and an aircraft. Airport operations and air traffic control is provided by the tax payer.
So we should create a federal communications authority, that simply goes ahead and lays the fiber. It should contract not with telcos but with the people who actually dig the trenches and splice the ends. Then they sell access at some nominal fee, not necessarily to make a profit or even break even but to avoid the tragedy of the commons. The resulting economic growth spurred by innovation in technology and new previously impossible business models then produces a tax base which indirectly pays for the infrastructure. This is really the ONLY way to get it done. Unfortunately our
With this technique you can cross-infect VMs and rootkit the hypervisor. That's a much bigger problem than simply rolling back a VM to a clean snapshot.
Is the industry good at picking stuff that becomes popular and sells? Or does it become popular because the industry tells us to like it and provides no alternative?
Music by its very nature is "sticky". If you hear something long enough you will become familiar with it and start liking it. (If not, then it's not musical.) The more similar it is to previously familiar music, the easier it sticks. What the major labels look for is music that is sufficiently similar to current mainstream to be cheap to make stick (say play five times over a week on radio, as opposed to 30 times over two months), while having at least some trivial differentiation and non-music appeal. Be it sexiness, dance skills, rebellion, or some other unrelated appeal.
Music doesn't spread by "word of mouth". It spreads by your friends playing it over and over while you're in their presence. It's a push process, not pull.
It's not surprising that teenagers and kids in their 20s download for free - because they can't afford to buy everything they've developed familiarity with. They lack the means to carry to the entertainment industry on their shoulders. And not having access to popular culture pretty much makes you a social outsider - not good for the development of teenagers, or the life of 20-somethings when so much hinges on social drama. Those of us who are older certainly pay for our music, but we can also afford to.
The popularity of a cultural work is largely a result not of any inherent qualities of the work itself, but of of the activities of the audience. ... This is a power law distribution with its long tail.
This applies to all aspects of human culture, not just music taste. This is why totalitarian thought control works: if you consistently eliminate or suppress the impopular 1-2% outliers on the curve the curve will narrow. But it retains it shape, so there is always an impopular fringe the vast majority might only miss "in principle". Principle being the classic liberal freedoms: of expression, equality before the law, absence of tyranny, and such abstract concepts. At least until integrity and principle itself is suppressed.
The record labels and the entertainment industry accomplish exactly the same result - whether wittingly or by coincidence. They target the tastes near the center of the curve and won't touch the outliers. Hence they produce a cultural narrowing; not many people would say they miss the outliers "because no one they know likes that crap anyway", other than perhaps in principle. The principle being that cultural plurality is inherently beneficial. The entertainment industry is good at producing choice, but with few alternatives. It's the same reason a walk down the supermarket bread isle reveals 30 kinds of breads, none with rye listed as a first ingredient, and only one or two without syrup, honey, or milk solids. Plurality of choice in the absence of alternative.
Powell's only sells books though. Only a tiny fraction of what I spend through Amazon is actually on books. I too would like a viable-one stop alternative.
When it comes to measurements, I'm much more interested in how long it takes from when you open the lid to when you can check you email or load a web page. So wake time, plus wifi auto connect time, plus DHCP config time, plus browser/mail connection and load/rendering time. Also, I'd be interested to know how long it will not only run on a charge, but also how long it will sleep on one, and whether I can swap batteries while it's sleeping and have it stay asleep. Assuming the battery is replaceable of course.
Most of the "performance" above is not hardware related at all, but software related, like how the wifi drivers and network protocols like DHCP are implemented, and how well the stack is vertically integrated. They're somewhat dependent on performance and capabilities provided by the hardware though.
For a laptop I think this is much, much, much more important than how well it plays some game.
DOCSIS is TDMA, which just doesn't play nice with TCP congestion control. This is why once the time slots become more intermittent as utilization goes up TCP performance tanks.
How can I create an NFS mount which will be mounted at boot time without a GUI?
Put it in /etc/fstab
See 'man fstab'
It doesn't really matter how much CO2 it emits. You grow something, then eat it, and then reemit it. The carbon then goes back to being absorbed by something growing - it's a closed system powered by the sun.
The increase in atmospheric carbon comes from us digging coal out of hillsides and pumping oil from underground, then burning it - adding it to the cycle. But it has nowhere to go in the short term, as in a few hundred or thousand of years, so results in increased atmospheric concentration.
Things that grow, live, eat, die, don't affect the net balance of atmospheric carbon. It doesn't matter how much you breathe, fart, your cows fart, or whether you brew beer. 1 lb of carbon in plant matter in can't result in more than 1 lb of carbon out or vice versa. It also makes no difference if it's CO2 or methane - the net result is for all practical purposes the same. (Methane is what, 15x more potent, but CO2 lingers about 18x longer; so over their lifetime it's a pound-for-pound toss-up.)
Likewise, if you ask a question of the form, "ARE zebras mammals?", in order to retrieve the answer, an algorithm that precisely represents an implementation of "Is this proposed assertion true" in the context of the specific data encoding must be run.
But for a SEARCH ENGINE, what you really want is not an answer to the question, but a list of results that are relevant to the reasoning involved in determining the answer.
The problem here is that the answer can vary depending on context, even a simple one like the one above. For instance, to everyday people zebras are Savannah Horsies, and horsies are mammals, so yes they _suppose_ this is correct. If you ask them, however, they might wonder if it's a trick question and may feel unsure. Biologically, zebras are mammals by taxonomy. Philosophically... what "is a mammal" (or taxonomy as such for that matter)? Is it a genetic pattern? Order of descent? Kinship through descent? Physical attributes? Is a non-mammal zebra, as a thought experiment, impossible? Is the mammal-ness merely a heuristic decision agreed to by a scientific quorum and if so could it be challenged?
What you might want from a search engine is anything related to answering the question - because presumably you're looking to deepen your understanding of something.
Because Steve Bourne was doing this work back in 1975! About that time people had only just got beyond programming by biting holes in paper-tape with their teeth!
There were certainly vastly more powerful command interpreters around. In fact, sh is a pretty crappy interface and was even then; its strength is in the ability to chain stdin/stdout and it simplified and cleaned up the system considerably by performing globbing. For a pure command line user interface, TOPS-20 was the gold standard in 1975. Sh's strength wasn't in its user friendliness, but in its scripting ability. Bash uses readline, but readline still lacks syntactic awareness.
Because the swiss have some sort of problem with a foreigner just waltzing into their country and taking a job when they've already got swiss that can do it.
As a scientist with a PhD and in your early 30s (i.e. with about 10 years' experience) you would:
1. Fly to Switzerland and stay up to 90 days without a visa if you're American/Canadian
2. Find employment in your field.
3. Have the employer obtain a specialist visa.
4. Work ten months.
5. Get an unlimited "C" work permit/visa.
With a "C" visa you're no longer tied to that particular employer but can move freely on the labor market.
Much easier than if you're an Indian PhD who wants to work in the U.S. - or a Swiss one for that matter. H1Bs are far more restrictive, and there is nothing like the Swiss "C" visa for when you're somewhere between H1B and permanent residency (at 48months IIRC). You also won't have to jump through silly administrative hoops, like go back to your home country to apply for the visa. And the Swiss will be nice, polite, and actually helpful at the border and immigration offices. As opposed to the snarling, incompetent, rude, bottom-of-the-barrel idiots you encounter here. If there's something wrong with your application or paperwork they will helpfully suggest how to correct it. Again, very different from U.S. government standards.
The fundamental flaw in the government's plan is this: in order for a wireless base station or home router to log, it has to log somewhere else. That means a computer with a static IP that must always be on.
Not really. My, by now rather antiquated, Netgear FVS318 firewall/NAT router will send a daily log by email. It can be set to log just about everything, including every connection made. If it has lots of logging to do it will send email more frequently. I sort this out using procmail and run it through a filter.
I propose if this bill passes people buy devices like this. Then set it to fully verbose logging and point it at: postmaster@gop.org. Problem solved - they get all the logs they could conceivably want and can safekeep their precious security hoard for themselves.
Nor does the government have the right to limit my consumption. The only things that should limit my consumption are my available resources and self control.
I don't think anyone cares about your consumption. We only care that you don't poison the village well - because if you do despite us repeatedly warning you not to, we'll string you up by your balls, cut you to little pieces, and use you for dog food.
Drive your SUV, but feed the tailpipe into the cabin and breathe your own toxic filth. Nobody cares if you do - just don't blow your crap into the same air the rest of us breathe.