For starters, OS' running on either virtual or simulated processors rather than physical ones won't necessarily use the physical instructions that have the vulnerabilities, no matter what the physical processor that the OS is technically using. (If I run Linux under ArcEm, and run ArcEm on an Intel processor, unless ArcEm itself uses the broken instructions, I cannot see how an attacker can reach the Intel processor from the Linux environment for the attack to take place. This is important because the composite environment is nothing more than a really heavy, multi-layer OS as far as the applications are concerned, and this attack is supposedly independent of OS.)
If it's via Java, then it must also depend some on the implementation. I doubt that IBM's java engine uses the same calls to the processor as Sun's, which means that there is further abstraction that the claim has to somehow deal with.
Now, on the opposite side of the argument, there's the issue of what happens if the claim is justified. If this is a remote exploit that is truly OS-independent, then it is a remote exploit that can hit OpenBSD, Trusted Solaris, and other secure OS'. These are OS' used for commercially-sensitive work and classified work. If they are potentially vulnerable to attack, that could seriously impact a lot of organizations that, well, really aren't going to like it. In the event of a conflict flaring up between Intel and the US Marines, we may see them moving the bombing practice areas for their aircraft into the North American mainland after all.
It shouldn't be hard.
on
Linux 2.6.26 Out
·
· Score: 2, Interesting
At least, to do a pre-configuration tool. You can currently list the processor(s), scan many of the physical and virtual busses, discover nearby network resources, etc. If that information were to be collected and then compared with "best practice" values for configuration options, it should be possible to set up many of the kernel options automagically. Not all, perhaps, but many. If you want something really sophisticated, you could do this as a two-pass thing. First pass uses "best practice" values and enables all of the kernel's profiling options. Second pass tweaks those values for the specific hardware being used using the profiling information, and disables all non-essential debug/profile options.
This would not build a "perfect" kernel, but it should produce something damn close, especially if the user is asked to supply information on anything that is uncertain or which cannot be deduced from the information that's discoverable or collectible.
There are also kernel patches, such as WEB100, which can tune some elements of kernel operation when a system is in use. If you add that into the mix, then you end up with something that is highly customized for the user without the user having to do more than the minimum of customization.
Here's a list of things Windows 7 MUST have, if people are to take it seriously:
True QoS, not just RSVP (which almost nobody uses and which doesn't scale)
True pre-emptible support (you need this for better multimedia)
True realtime support (you need this for better gaming)
True inter-process communication (Microsoft's RPC and DCOM don't cut it)
True low-cost threading
True host security using mandatory access controls and role-based security models (B3 or better, on the Orange Book scale)
True network security (Microsoft has more programmers than the OpenBSD team, so can audit more code more thoroughly)
True logging filesystem (journaling doesn't cut it any more with data centers)
True carrier-grade reliability (though better is always nice)
True object-oriented desktop (associations by file extension? CP/M called and wants its ideas back)
True high-performance networking (where's Microsoft's WEB100? you seen much VNIC or RNIC support either?)
True clustering (sorry, recompiling MPICH and calling it Windows Cluster Edition doesn't cut it)
Just 12 really trivial features that would boost Windows in the gaming, multimedia, enterprise, data center and scientific markets. 12 lousy features. Not even the 20 that TFA asked for. Just 12. 12 features that other operating systems have achieved with a fraction of the resources available to them. 12 features that should not be hard to add, and require no concessions to be made to other OS designs or other architectures. 12 features that don't hurt the prevailing egos or require any kind of shift in overall corporate philosophy.
12 additions we are unlikely to ever see in Windows.
I'd rather they ask Slashdot than Microsoft, Google or Yahoo. However, it would probably be just as useful to ask on I Can Has Cheezburger or Cute Overload. (OMG, Ponies!!!)
Depends on what you mean by similar, for a start. I would not define a planet according to where it happens to be in the solar system, but rather according to composition, structure and mass, as these are things which we know for a fact to distinguish planets from asteroids (eg: asteroids have no core) and planets from comets (eg: comets have multiple cores). I would define a new class for objects for which insufficient data existed to produce a firm classification, but that is it.
Why does it matter? Well, think back a few days to the recent news on the DNA analysis of birds. Turns out, the definition based on appearances is completely wrong. What was it, kestrels are genetically closer to hummingbirds than any other bird of prey? And the DNA variation between any two lineages within a species has next to zero correspondence to morphology. In other words, looking at something from the outside tells you bugger all. So, naturally, looking at the outside of an object orbiting the sun is the perfect way to tell what it is. It's only a method every other discipline has now ruled to be faulty, after all.
Some - but not all - forms of autism have nothing to do with "not thinking". Rather, it has to do with thinking too much. There's an excellent autobiographical series, of which the second is "Somebody, Somewhere", which illustrates that certain forms of autism are caused by a collapse in the internal divisions in the brain, that by reducing sensory input to manageable levels, these people are perfectly fine. (And, yes, the example I'm thinking of would be what is commonly diagnosed as Low Functioning Autism.) If the data inputs are at "normal" background level, such minds are swamped and shut down from overload.
This would obviously not be good, if we expected everyone to be 100% independent, rather than interdependent. If you're interdependent, a highly specialized brain that is perfectly tuned to a narrow range of things will work. In these cases, such people will be able to excel at those things their brain is tuned for, much as a games machine and a supercomputer excel at their specialties but would fail totally at trying to do the other's tasks.
This is NOT the same as "idiot savant"-style gifts, where there is no real processing involved. but it is connected in that these are minds capable of greater attention to detail and greater precision than any "normal" person. And because the walls in the mind have collapsed, they should be capable of connecting data together well beyond what you or I could do.
But there are other forms. Autism from Fragile X will be different from autism from other causes, for example. Some of these forms of autism may very will shut down thinking totally, rather than just when there's too much data. These forms of autism would not offer any obvious advantage to the person as far as I can tell, but I am willing to accept that there is a possibility that they do, somehow, and will not allow my personal belief in the supremacy of the intellect to overrule the rather obvious fact that I can logically invalidate other people's just-as-strong beliefs in the supremacy of their ideals. If they can be wrong, then so can I, and I have no more right to inflict my values on others than they have to inflict their values on me.
Does that mean that if you can demonstrate - beyond any shadow of a doubt - that a person is suffering, that they would/do not want to suffer, and that they gain no benefit whatsoever from their condition, that I would insist that they continue to suffer? No. That would be stupid, malicious or both. That may well be the case for your cousin, and if so, I hope that that specific instance can be cured. The problem is, from just the vastly overused label of autism, I cannot possibly tell that.
I'll agree. Many mutations have advantages as well as disadvantages, including the mutation known as "the average joe" - who is probably more disadvantaged than anyone else. In the same way that coders often use the maxim of "speed, size and simplicity - pick any two", the same is likely true of many of the variants found in human DNA. I would be extremely wary of allowing insurance agencies, jobs, or "social norms" to decide which variants were acceptable and which needed to be fixed.
(Many aspies hate and revile organizations who consider them to be lesser beings who should be "cured", whether we want it or not. Yes, some do advocate cures against the will of the one being "cured". I think such organizations and such attitudes are an abomination and far more in need of "curing" than Asperger's or Autism.)
Do we want human evolution - which has actually been accelerating over the past 10,000 years - to come to a complete stop? Are we willing to face the only possible consequence of such an event (extinction)? Are the fragile egos of a few corporate executives worth that much to us, as a species? The variation in human DNA is very close to the difference between the reference DNA of humans and the reference DNA of chimps. (The absolute percentage is of no consequence, if the variation means there is a potential of overlap, and I'm not interested here in whether such overlap exists or is merely approached.) If we start "fixing" DNA, how much of that variation do we condemn to oblivion? And can we be oh so amazingly certain that the variants we so condemn aren't exactly the variants we need?
(Think Black Death. The mutation that increased resistance to Bubonic Plague decreased resistance to other diseases that are now proving fatal today, such as Ebola and Marburg. Those most likely to survive the modern killers are least likely to survive those diseases we can now cure by other means. The "reference" DNA is now the broken copy, the unpatched version has better survival odds. But those obsessed with "fixing" those of us who are "broken" would have it the other way round. You MUST apply the patch, or suffer serious social consequences, even if it means you are at greater risk of dying or being a contributor to the death of many. Conformism is bloody dangerous and should be outlawed.)
Generally, there are two distinct characteristics involved - structure (syntax, grammar, etc) and bandwidth (the scope of the information that can be delivered, and how long it takes to deliver it). A dog can communicate, but the structure is minimal and the bandwidth is - frankly - pathetic. However, it's quite sufficient for the purposes of hunters trying to coordinate a large pack for a successful foray. Humans have very complex speech patterns requiring elaborate structure and extremely high bandwidth, but it delivers rather more than the next meal. Technology, art and abstract thought would be impossible without such complexities, which is why it is so key to understanding human development. (It is also key in understanding non-human development in all animals that exhibit complex language and/or complex behaviour, since the better our understanding of how these connect, the better our understanding of what is taking place in societies we cannot readily communicate with and the better able we will be to guague what is real and what is mere anthromorphising.)
Factor in that people can (and do) invent their own written languages. I never mastered writing longhand notes at University (verbal and written information was different), so I developed my own written shorthand. I expect some cave paintings are likewise early human shorthand, teaching about hunting rather than being magical attempts for achieving hunting success. If that is correct, then crude proto-writing existed for a long time, but was never taken anywhere, and the consensus is that it is not writing.
That's much the way writing evolved (pictograms evolving into abstract representations, the representations then moving from physical objects to an abstract concept and finally to a sound that could double-up as that concept, which led to true writing as new concepts could then be composed from combining the archetypes together, and so on). The idea that language as a whole followed the same basic evolutionary path as writing is not that far-fetched. I'm actually rather surprised that the oldest known true writing (was Sumerian, the current record-holder is a form of ancient Chinese, dating 3,000 years ealier) is many hundreds of thousands of years newer than language, as things like art (eg: the flute, early necklace beads, etc) and symbolism (eg: the earliest known examples of abstract ritual) are hard to transmit between individuals by example alone. Writing is amazingly modern, in comparison to either cause or need.
The holes in the Neanderthal bone flute were carved (no internal fracturing or splintering, as you would expect from an animal bite) and regional variations in Neanderthal tools in Britain have suggested the possibility of regional culture at a very early date. These have long hinted at language being a much earlier development than believed. This adds a lot of weight to the argument, but it is the fact that there are an overwhelming number of pointers and indicators for language that should clinch it. Studies on hominids that far back is inherently speculative, which means those doing the studying have to carefully examine evidence with a skeptical eye. As a result, no one discovery will ever cause a radical shift in and of itself, but radical shifts - when they happen - will be all the more stunning.
...for a router to require rebooting are memory leaks (especially for the routing tables or ARP tables), buffer overflows (same), a portscan or other attack - say by a zombie or skriptkiddie - putting the system into an unrecoverable state (eg: resources exhausted), or a kernel (likely driver) bug putting the kernel into an unrecoverable deadlock. There's almost nothing else that can possibly go wrong in a software router, at least to the point of locking the system up.
Ok, the router software - likely ripd, xorp, quagga or zebra for any domestic ADSL router - might crash, but the worst that will happen then is that you don't learn new routes. Since DSL providers don't tend to switch their internal IP addresses very often, that should not impact any existing subnet. It means tunnels can't be generated on-the-fly, it also means your next-door neighbor can't connect their LAN party to your wireless connection, but it shouldn't impact you in the slightest.
The next question, however, is how on Earth are you noticing the router needs rebooting? The kernel is quite capable of rebooting itself under many (but not all) soft lockups. Linux provides several such mechanisms for doing just that. A simple watchdog circuit, using a bistable circuit, a couple of capacitors, a relay and a trigger line that has to change state, could be added by a manufacturer for maybe a couple of dollars. It probably doesn't even need to be that complex.
When it does reboot, LinuxBIOS is under 3 seconds and I don't imagine OpenBIOS is that much slower. Intel's Tiano probably is, but it's open source so you can rip out anything that's useless. Therefore, recovery times should be barely detectable to an end user. (Most websites vary in download times by more than 3 seconds between visits. Unless you're playing Netrek or WoW at that precise moment, I seriously doubt you'll notice a 3 second outage.)
Finally, however, why isn't the router using carrier-grade software? Again, carrier-grade Linux exists, which should give you 5N uptimes in the worst possible case. Domestic routers are not worst-possible. Even data centers rarely get the kind of stress that could be expected to force an unrecoverable state. If your router is not overheating, has plenty of RAM, and needs rebooting more than once every other year, there is something seriously defective in the software or hardware.
Well, direct solar heating (say, for water) is much more efficient than solar electrical generation that then powers an electrical heater. Since (essentially) everyone consumes hot water, it would make much more sense to directly generate, then store, the desired end result than generate multiple intermediate steps that are less efficient and more expensive. The reduction in demand for electricity should rapidly produce greater savings than the cost of setting up such a distributed network. (Solar heaters are so efficient, they're even used in Wales, not a country known for masses of sunlight.)
There's also the matter of nuclear fusion. Yes, there isn't a working design yet, but if you invested enough into getting one, that's just a matter of time. We already have an idea of what the exterior superstructure would need to be, in terms of scale and design. Given that just building that would take many years, there's no reason not to do that in parallel.
Lastly, there's one small issue I have with these superconducting cables. 77K? There are superconductors that operate at something like 2-2.5 times that temperature! Are these guys wanting to prove it can't be done?
If - and this is a big if - the current model for galaxy formation is wrong, then this could have all kinds of consequences to cosmological theories. For example, the current estimate for the amount of dark matter (or, indeed, the need for it) is based on what would be needed for galaxies to work under the current models. There are a lot of interdependencies in cosmology. Another consequence is that they'll need to revisit estimates for the number of rogue stars that lie outside of galaxies. Given the frequency of galactic collisions, and given the new, revised estimate for star formation, you should expect to find a lot of rogue stars maybe a billion light-years closer than this galaxy.
I always thought the industry was too conservative to allow chips to Flash. Seriously, this is going to have serious potential in the solid-state disk arena, but will probably not affect firmware and BIOSes as you don't update or access those a vast amount. For SSDs, you really want this for the bulk storage but battery-backed RAM for transaction logging and caching as you don't need those to be longer-lived than the time it needs to complete a full transaction.
(Ideally, you'd have battery-backed "smart RAM" that can complete transactions into memristor storage even if disconnected from a main computer and power supply. Then you'd have something that was not only fast but proof against most idiots. Not all, as idiots are so inventive, but most.)
Oh, if there's one thing Trimet is superb at, it is talking. I've yet to see any upgrading of ticket machines - they were out again today, fixing one of the Win 3.11 boxes at the Rose Quarter - and with the promised price hikes in September to cover fuel costs, I doubt they have much to throw into upgrading. (What would they upgrade to, Vista? Can you imagine what's going to happen when customers are asked to confirm that they want the ticket machine to have permission to print the ticket?)
There is an example of systems failure causing the loss of a ship - although I do not believe Microsoft was at fault. (I'll blame them anyway, to be consistent.) That example was HMS Sheffield, in the Falkland's War, which was hit by an exocet missile despite having the ability to shoot them down. The point defense systems were confused by too many objects on the RADAR.
That blunder in systems design cost lives. A great many lives. Totally needlessly. Don't imagine it can't happen to the US navy, because if they rely on unstable software on mission critical systems, it will.
Another non-Microsoft example of why software should be treated with a bit more care was the Boeing 767 that "landed" at Heathrow after all onboard computers shut down in flight. The pilots were damn good and damn lucky, but luck aside, why the hell were there no backup computer systems or failover strategies? Why did the pilots have to "fly" with no engines, no instrumentation and very nearly no controls?
But when you combine this kind of insanely poor systems design with Microsoft's unreliability and long boot times, you have something that is asking for trouble. Problem is, if you ask for trouble nicely enough, trouble is happy to oblige.
The local mass transit (Trimet) ticketing system runs entirely on Windows 3.11. Found that out when one of the maintenance engineers was rebooting one of the systems. Apparently, the machines are imported, Trimet doesn't have permission to do any software maintenance, there's no way in hell they're going to be able to afford to pay German engineers to come to the west coast to do a software update at all those locations, and there's an even smaller chance Trimet is going to be happy running multiple versions of the software, as it means getting engineers with greater skills (which will cost) and they'll have to keep a wider range of spare parts (which will also cost).
I could very easily see them buying machines that are not technically licensed from Microsoft, on the grounds that Microsoft lawyers don't ride light rail, a little fudging of dates would conceal it from any realistic audit, and replacing every single kiosk with one that is powerful enough to run Vista would be insanely expensive both to buy and to run (electricity isn't free).
IPSec encrypts everything, making it much harder for an eavesdropper to identify when someone is transmitting something that may be of interest. It also authenticates the recipient, making man-in-the-middle attacks much harder. We saw with the recent DNS scare how easy it would have been for anyone - let alone a three-letter agency - to redirect traffic through a proxy of their choosing, making it trivial to perform MitM attacks on any vulnerable encryption method.
I would also suggest - when network performance isn't critical - relaying traffic over the 6Bone to any gateway (if using 4-over-6) or a 6-to-4 proxy to drop back onto the regular Internet at a remote location. I sincerely doubt the 6Bone is monitored to any real degree, which means you can also obscure your point of origin. (IPv6 Phase 2 seems to be going well, incidentally, with 153 products listed as gold-certified and 352 silver-certified. However, it is still reasonably "unknown", making this an ideal time to learn how to use it to improve anonymity and privacy. You do NOT want to end up playing catch-up on such things.)
Tor is also good, and I would definitely recommend that Tor be used in a similar way - as a backbone that cannot be realistically monitored, thus providing extra layers of anonymity and privacy that a simple, direct, encrypted stream cannot.
If you do use direct connections, though, be sensible about the encryption methods. Use the strongest encryption mode you can, not just the strongest encryption key. A good key is important, but information can be extracted from a strongly encrypted message if you are using ECB or some other very weak encryption mode. There are now very strong modes, although not all encryption software or hardware support them. If they are supported both ends, though, use them.
I would agree with you. Especially the snide suggestion that there are no GPL coders who could re-implement the JPEG library. Firstly, that was uncalled-for. Secondly, if any coder saw any value in such a re-implementation under the GPL, you can bet there'd be a dozen already. You only need to look at the number of crypto libraries to realize there are plenty of coders who are perfectly competent at advanced mathematics. It is true that I know of many more maths libraries that are BSD licensed than GPLed, but most of those libraries I know of are either paid for by commercial operations (where there is a vested interest in closable code) or written by "Red Brick" Universities, where it has probably taken this long to get the senior professors to understand the BSD license.
Slashdot Mirror
If it's via Java, then it must also depend some on the implementation. I doubt that IBM's java engine uses the same calls to the processor as Sun's, which means that there is further abstraction that the claim has to somehow deal with.
Now, on the opposite side of the argument, there's the issue of what happens if the claim is justified. If this is a remote exploit that is truly OS-independent, then it is a remote exploit that can hit OpenBSD, Trusted Solaris, and other secure OS'. These are OS' used for commercially-sensitive work and classified work. If they are potentially vulnerable to attack, that could seriously impact a lot of organizations that, well, really aren't going to like it. In the event of a conflict flaring up between Intel and the US Marines, we may see them moving the bombing practice areas for their aircraft into the North American mainland after all.
This would not build a "perfect" kernel, but it should produce something damn close, especially if the user is asked to supply information on anything that is uncertain or which cannot be deduced from the information that's discoverable or collectible.
There are also kernel patches, such as WEB100, which can tune some elements of kernel operation when a system is in use. If you add that into the mix, then you end up with something that is highly customized for the user without the user having to do more than the minimum of customization.
Just 12 really trivial features that would boost Windows in the gaming, multimedia, enterprise, data center and scientific markets. 12 lousy features. Not even the 20 that TFA asked for. Just 12. 12 features that other operating systems have achieved with a fraction of the resources available to them. 12 features that should not be hard to add, and require no concessions to be made to other OS designs or other architectures. 12 features that don't hurt the prevailing egos or require any kind of shift in overall corporate philosophy.
12 additions we are unlikely to ever see in Windows.
I'd rather they ask Slashdot than Microsoft, Google or Yahoo. However, it would probably be just as useful to ask on I Can Has Cheezburger or Cute Overload. (OMG, Ponies!!!)
Why does it matter? Well, think back a few days to the recent news on the DNA analysis of birds. Turns out, the definition based on appearances is completely wrong. What was it, kestrels are genetically closer to hummingbirds than any other bird of prey? And the DNA variation between any two lineages within a species has next to zero correspondence to morphology. In other words, looking at something from the outside tells you bugger all. So, naturally, looking at the outside of an object orbiting the sun is the perfect way to tell what it is. It's only a method every other discipline has now ruled to be faulty, after all.
Stop throwing things at it, then.
Are you using BSD make or GNU make?
This would obviously not be good, if we expected everyone to be 100% independent, rather than interdependent. If you're interdependent, a highly specialized brain that is perfectly tuned to a narrow range of things will work. In these cases, such people will be able to excel at those things their brain is tuned for, much as a games machine and a supercomputer excel at their specialties but would fail totally at trying to do the other's tasks.
This is NOT the same as "idiot savant"-style gifts, where there is no real processing involved. but it is connected in that these are minds capable of greater attention to detail and greater precision than any "normal" person. And because the walls in the mind have collapsed, they should be capable of connecting data together well beyond what you or I could do.
But there are other forms. Autism from Fragile X will be different from autism from other causes, for example. Some of these forms of autism may very will shut down thinking totally, rather than just when there's too much data. These forms of autism would not offer any obvious advantage to the person as far as I can tell, but I am willing to accept that there is a possibility that they do, somehow, and will not allow my personal belief in the supremacy of the intellect to overrule the rather obvious fact that I can logically invalidate other people's just-as-strong beliefs in the supremacy of their ideals. If they can be wrong, then so can I, and I have no more right to inflict my values on others than they have to inflict their values on me.
Does that mean that if you can demonstrate - beyond any shadow of a doubt - that a person is suffering, that they would/do not want to suffer, and that they gain no benefit whatsoever from their condition, that I would insist that they continue to suffer? No. That would be stupid, malicious or both. That may well be the case for your cousin, and if so, I hope that that specific instance can be cured. The problem is, from just the vastly overused label of autism, I cannot possibly tell that.
(Many aspies hate and revile organizations who consider them to be lesser beings who should be "cured", whether we want it or not. Yes, some do advocate cures against the will of the one being "cured". I think such organizations and such attitudes are an abomination and far more in need of "curing" than Asperger's or Autism.)
Do we want human evolution - which has actually been accelerating over the past 10,000 years - to come to a complete stop? Are we willing to face the only possible consequence of such an event (extinction)? Are the fragile egos of a few corporate executives worth that much to us, as a species? The variation in human DNA is very close to the difference between the reference DNA of humans and the reference DNA of chimps. (The absolute percentage is of no consequence, if the variation means there is a potential of overlap, and I'm not interested here in whether such overlap exists or is merely approached.) If we start "fixing" DNA, how much of that variation do we condemn to oblivion? And can we be oh so amazingly certain that the variants we so condemn aren't exactly the variants we need?
(Think Black Death. The mutation that increased resistance to Bubonic Plague decreased resistance to other diseases that are now proving fatal today, such as Ebola and Marburg. Those most likely to survive the modern killers are least likely to survive those diseases we can now cure by other means. The "reference" DNA is now the broken copy, the unpatched version has better survival odds. But those obsessed with "fixing" those of us who are "broken" would have it the other way round. You MUST apply the patch, or suffer serious social consequences, even if it means you are at greater risk of dying or being a contributor to the death of many. Conformism is bloody dangerous and should be outlawed.)
Generally, there are two distinct characteristics involved - structure (syntax, grammar, etc) and bandwidth (the scope of the information that can be delivered, and how long it takes to deliver it). A dog can communicate, but the structure is minimal and the bandwidth is - frankly - pathetic. However, it's quite sufficient for the purposes of hunters trying to coordinate a large pack for a successful foray. Humans have very complex speech patterns requiring elaborate structure and extremely high bandwidth, but it delivers rather more than the next meal. Technology, art and abstract thought would be impossible without such complexities, which is why it is so key to understanding human development. (It is also key in understanding non-human development in all animals that exhibit complex language and/or complex behaviour, since the better our understanding of how these connect, the better our understanding of what is taking place in societies we cannot readily communicate with and the better able we will be to guague what is real and what is mere anthromorphising.)
Factor in that people can (and do) invent their own written languages. I never mastered writing longhand notes at University (verbal and written information was different), so I developed my own written shorthand. I expect some cave paintings are likewise early human shorthand, teaching about hunting rather than being magical attempts for achieving hunting success. If that is correct, then crude proto-writing existed for a long time, but was never taken anywhere, and the consensus is that it is not writing.
That's much the way writing evolved (pictograms evolving into abstract representations, the representations then moving from physical objects to an abstract concept and finally to a sound that could double-up as that concept, which led to true writing as new concepts could then be composed from combining the archetypes together, and so on). The idea that language as a whole followed the same basic evolutionary path as writing is not that far-fetched. I'm actually rather surprised that the oldest known true writing (was Sumerian, the current record-holder is a form of ancient Chinese, dating 3,000 years ealier) is many hundreds of thousands of years newer than language, as things like art (eg: the flute, early necklace beads, etc) and symbolism (eg: the earliest known examples of abstract ritual) are hard to transmit between individuals by example alone. Writing is amazingly modern, in comparison to either cause or need.
The holes in the Neanderthal bone flute were carved (no internal fracturing or splintering, as you would expect from an animal bite) and regional variations in Neanderthal tools in Britain have suggested the possibility of regional culture at a very early date. These have long hinted at language being a much earlier development than believed. This adds a lot of weight to the argument, but it is the fact that there are an overwhelming number of pointers and indicators for language that should clinch it. Studies on hominids that far back is inherently speculative, which means those doing the studying have to carefully examine evidence with a skeptical eye. As a result, no one discovery will ever cause a radical shift in and of itself, but radical shifts - when they happen - will be all the more stunning.
Ok, the router software - likely ripd, xorp, quagga or zebra for any domestic ADSL router - might crash, but the worst that will happen then is that you don't learn new routes. Since DSL providers don't tend to switch their internal IP addresses very often, that should not impact any existing subnet. It means tunnels can't be generated on-the-fly, it also means your next-door neighbor can't connect their LAN party to your wireless connection, but it shouldn't impact you in the slightest.
The next question, however, is how on Earth are you noticing the router needs rebooting? The kernel is quite capable of rebooting itself under many (but not all) soft lockups. Linux provides several such mechanisms for doing just that. A simple watchdog circuit, using a bistable circuit, a couple of capacitors, a relay and a trigger line that has to change state, could be added by a manufacturer for maybe a couple of dollars. It probably doesn't even need to be that complex.
When it does reboot, LinuxBIOS is under 3 seconds and I don't imagine OpenBIOS is that much slower. Intel's Tiano probably is, but it's open source so you can rip out anything that's useless. Therefore, recovery times should be barely detectable to an end user. (Most websites vary in download times by more than 3 seconds between visits. Unless you're playing Netrek or WoW at that precise moment, I seriously doubt you'll notice a 3 second outage.)
Finally, however, why isn't the router using carrier-grade software? Again, carrier-grade Linux exists, which should give you 5N uptimes in the worst possible case. Domestic routers are not worst-possible. Even data centers rarely get the kind of stress that could be expected to force an unrecoverable state. If your router is not overheating, has plenty of RAM, and needs rebooting more than once every other year, there is something seriously defective in the software or hardware.
There's also the matter of nuclear fusion. Yes, there isn't a working design yet, but if you invested enough into getting one, that's just a matter of time. We already have an idea of what the exterior superstructure would need to be, in terms of scale and design. Given that just building that would take many years, there's no reason not to do that in parallel.
Lastly, there's one small issue I have with these superconducting cables. 77K? There are superconductors that operate at something like 2-2.5 times that temperature! Are these guys wanting to prove it can't be done?
Obviously DARPA and the Department of Energy are commie mutant traitors, requiring DHS oversight. That, and I should cut down on playing Paranoia.
If - and this is a big if - the current model for galaxy formation is wrong, then this could have all kinds of consequences to cosmological theories. For example, the current estimate for the amount of dark matter (or, indeed, the need for it) is based on what would be needed for galaxies to work under the current models. There are a lot of interdependencies in cosmology. Another consequence is that they'll need to revisit estimates for the number of rogue stars that lie outside of galaxies. Given the frequency of galactic collisions, and given the new, revised estimate for star formation, you should expect to find a lot of rogue stars maybe a billion light-years closer than this galaxy.
Isolinear? Isolinear??! When you could be running Tarriel Cell technology?
(Ideally, you'd have battery-backed "smart RAM" that can complete transactions into memristor storage even if disconnected from a main computer and power supply. Then you'd have something that was not only fast but proof against most idiots. Not all, as idiots are so inventive, but most.)
Oh, if there's one thing Trimet is superb at, it is talking. I've yet to see any upgrading of ticket machines - they were out again today, fixing one of the Win 3.11 boxes at the Rose Quarter - and with the promised price hikes in September to cover fuel costs, I doubt they have much to throw into upgrading. (What would they upgrade to, Vista? Can you imagine what's going to happen when customers are asked to confirm that they want the ticket machine to have permission to print the ticket?)
The correct modern spelling is kitteh, and you apparently solve that problem by supplying them with elebenty cheezburgers.
That blunder in systems design cost lives. A great many lives. Totally needlessly. Don't imagine it can't happen to the US navy, because if they rely on unstable software on mission critical systems, it will.
Another non-Microsoft example of why software should be treated with a bit more care was the Boeing 767 that "landed" at Heathrow after all onboard computers shut down in flight. The pilots were damn good and damn lucky, but luck aside, why the hell were there no backup computer systems or failover strategies? Why did the pilots have to "fly" with no engines, no instrumentation and very nearly no controls?
But when you combine this kind of insanely poor systems design with Microsoft's unreliability and long boot times, you have something that is asking for trouble. Problem is, if you ask for trouble nicely enough, trouble is happy to oblige.
I could very easily see them buying machines that are not technically licensed from Microsoft, on the grounds that Microsoft lawyers don't ride light rail, a little fudging of dates would conceal it from any realistic audit, and replacing every single kiosk with one that is powerful enough to run Vista would be insanely expensive both to buy and to run (electricity isn't free).
I would also suggest - when network performance isn't critical - relaying traffic over the 6Bone to any gateway (if using 4-over-6) or a 6-to-4 proxy to drop back onto the regular Internet at a remote location. I sincerely doubt the 6Bone is monitored to any real degree, which means you can also obscure your point of origin. (IPv6 Phase 2 seems to be going well, incidentally, with 153 products listed as gold-certified and 352 silver-certified. However, it is still reasonably "unknown", making this an ideal time to learn how to use it to improve anonymity and privacy. You do NOT want to end up playing catch-up on such things.)
Tor is also good, and I would definitely recommend that Tor be used in a similar way - as a backbone that cannot be realistically monitored, thus providing extra layers of anonymity and privacy that a simple, direct, encrypted stream cannot.
If you do use direct connections, though, be sensible about the encryption methods. Use the strongest encryption mode you can, not just the strongest encryption key. A good key is important, but information can be extracted from a strongly encrypted message if you are using ECB or some other very weak encryption mode. There are now very strong modes, although not all encryption software or hardware support them. If they are supported both ends, though, use them.
I would agree with you. Especially the snide suggestion that there are no GPL coders who could re-implement the JPEG library. Firstly, that was uncalled-for. Secondly, if any coder saw any value in such a re-implementation under the GPL, you can bet there'd be a dozen already. You only need to look at the number of crypto libraries to realize there are plenty of coders who are perfectly competent at advanced mathematics. It is true that I know of many more maths libraries that are BSD licensed than GPLed, but most of those libraries I know of are either paid for by commercial operations (where there is a vested interest in closable code) or written by "Red Brick" Universities, where it has probably taken this long to get the senior professors to understand the BSD license.