Selling private/personal/confidential information in an effort to destroy or cause definite harm to the third party concerned. (eg: cops who sell info on informants to the Mafia)
Providing sensitive information for the express purpose of causing economic harm for political gain (eg: The classic example was where the NSA wiretapped Airbus' commercial activity and sold the data to Boeing)
Publishing names and addresses of witnesses, particularly in trials involving large-scale corruption or organized crime
These are all cases where the truth should, very reasonably, be of limited availability. They are, I acknowledge, fairly extreme examples but that is because for the most part it really takes an extreme case to justify such an extreme measure. (As you note, blackmail is also such a case - I guess it's still extreme for the individual concerned, though.) In my examples, I also note the harm involved, so I guess the second constraint is when intended harm exceeds good. All of the parent's examples and mine also concern themselves with speech that is not "freeing" or "unprejudiced", but is for the express purpose of "control" and "manipulation" - either of the target audience and/or, where different, the targets of the speech.
Where a reasonable person can conclude that the speech is within such an extreme circumstance that the consequences will be disproportionate to the speech itself, AND where that speech is intended to cause disproportionate harm, AND where that speech is not truly "free" but laced with spin, THEN I can see no sensible objection to constraint on that speech. In fact, it better well had be constrained*.
Where a reasonable person can conclude that the speech is in a moderate or benign contect, AND that the speech is intended to bring about good, AND where the speech is not only truthful but devoid of spin or propoganda, THEN that speech should be protected utterly and unconditionally.
This leaves all other permutations. I would argue that as you approach the absolute of the extreme case, you will get fewer and fewer cases that are clear-cut cases of where speech should be unconditionally free. However, I believe absolutely that the assumption should be one of speech being free and that those who would limit it should prove unconditionally and beyond all reasonable doubt that the listed factors are present and that they outweigh all other factors.
*Where a person or an agency deliberately and knowingly sets themselves up in a situation where exposure would seriously discredit that person or agency - particularly if it is a crime, a judge should have the power to rule that they are not entitled to legal protections from the consequences of their own actions.
Under NO circumstances should a Government be able to tell a court what should be secret and what shouldn't. I do not believe in such a right and see no justification for it, if only for the simple reason that that violates the separation of powers and also creates a conflict of interest.
Emacs can have macros, run shell scripts, etc. DEC VMS' default editor - EVE - supported DCL (DEC Command Language) script. uMicro's OS (ghastly as it was) was fully object-oriented, in that everything was an object and you could run whatever methods you liked on that object.
I absolutely love playing with new technology - can't get enough arcane, bizare and downright weird programs that do stuff that's novel or just plain strange. I hope ACME fits into this category, but as the above list shows, it has tough cometition before it qualifies as new & interesting (at least to me). Being able to store scriptlets in one window to apply to another might qualify, if there's some new tangent to it. Oh, and I'd have to be sure that the method used to apply scripts in this way did not pose a security issue -- the vast majority of all the viruses currently for Windows are macro viruses, and the early (AT&T) history of Unix includes tales of viral backdoors.
Trust me, I want to be convinced, if for no other reason than I'm running out of new programs to play with. The nightmare of withdrawl symptoms, suffering from stale sameness... It doesn't bear thinking about!
Folding editors would be great - folding word processors would also be cool for much the same reason. And, yes, I remember Occam and the Transputer! Wonderful devices, sadly neglected. But all is not lost, for Kent University has an Occam compiler (KROC) for *nix boxes (Linux, the BSDs, etc). The documentation sadly proves that (code skill) x (documentation skill) = constant. However, it really blows the socks off any other programming language out there, feature-wise.
I don't know if KROC has been ported to Plan9/Inferno yet, but it damn well should be.
With the move away from US Government-funded infrastructure towards a purely profit-making attitude, virtually any redundancy in the Internet has been eroded at best, eliminated at worst. Redundancy costs hard cash and earns nothing extra. The days of the backbone being able to survive a full-scale nuclear attack are over. These kinds of attacks will persist - and worsen - because an individual is quite capable of summoning a cyber-army of zombies that can easily take out any one of a number of single points of failure.
The backbone providers are unlikely to care that much - it impacts a little business, but most make money off their inter-corporate and inter-Governmental lines. The more the Internet degrades, the more high-priced services the major vendors can sell and the more copper/fiber the telecos can charge for. I don't see much of a motive to fix things here.
The vendors further up the chain don't need to care much, either. The companies on the Internet can't gain by switching ISP, because it's the backbone that's broken and they'll have to go through it to reach the peasents - err, home users anyway. The corporations that sell over the Internet don't lose any sales, as a person who is going to buy from an online store is likely to be doing other stuff and won't go out to the stores, so they'll be back. Home users, for the most part, are ignorant enough to think AOL and MSN are really neat ideas, have no clue what the Internet involves, what needs fixing or why, and is likely to pass it off as someone else's problem anyway. And those who ARE smart enough are Libertarian enough that they won't Unionize and DEMAND the fixes that damn well should be made.
(IT users and IT professionals should stop with the "unions are evil" crap - no organization is any more evil than the people in it - and collectively insist that the defects be fixed. No ifs, no buts, no maybes, no excuses, no delays - these kinds of attacks SHOULD be impossible and COULD - very cheaply - be made impossible. But nobody is going to even take the cheap option without a fight, if there's an even cheaper option of apathy open to them.)
Well, it's true that a lot of places I've seen use SNMP v1 or v2, which aren't particularly secure, but avoid using SNMP v3, which is. No company I've worked for has the skillbase to install IPSec, never mind direct their system health and maintenance traffic over it. You're right that a few thousand miles is a bit of a distance to run to reboot a server - although they do have a five-year-old who would probably be willing to give it a try. (There's a very young kid there who can apparently sprint double marathon distances, daily. The Government there is trying to stop his coach from killing him through overwork.)
Given that corporations have been very careful to put their primary servers inside the corporate firewall to protect them, blasting the system load and application status over the public internet would seem to be retrograde. If there's a "Big Brother" or MRTG server in the DMZ that can see the internal servers, then there's a machine crackers can use to bypass all internal security measures. And what are these Indian IT guys going to use to log in, anyway? There's a frightening possibility that connections will be through raw telnet or the most basic setting on Terminal Services, and that corporations will simply plain-text e-mail the usernames and passwords required to the Indian company managing their systems.
All this ignores the bandwidth issue. Most companies can barely afford to get the pipes they need. If you're now going to fire RRD databases and SNMP streams over to India, you're going to eat into that bandwidth. As soon as you throw in virtual consoles for GUI management (and that's the most likely way they'd control most of the boxes), you're talking severe network hogs. This means that corporations with insufficient bandwidth will kill their connections entirely or have to spend more on additional bandwidth than they're saving by outsourcing.
The Government is apparently trying to get the evidence quashed independently, claiming state secrets priviledge. (The Wired article claims that this comes from UK Common Law, but UK common law comes from the Magna Carta and the Magna Carta made no such provision. Indeed, it stated clearly that nobody could be denied the right to justice, and that courts were forbidden from ruling on the basis of a single person's unsupported testimony, which is what a secrecy order without proof would be.)
In the same way that a trade secret that becomes public ceases to be protectable as a trade secret, I would have though that this would cease to merit any protections as it is self-evidently no longer secret, whatever the state may say.
So, on the basis that state secrets does NOT appear to be a valid piece of Common Law, and that there is no secret left to protect, I can see no justification for quashing this evidence. Furthermore, as the documents HAVE been published openly, AT&T have lost all rights to their claim of trade secrets, and so I can see no obvious justification of the evidence even being sealed. We already know what the bulk of it says, as it's online!
The argument over who is right and who is wrong is, in this case, largely academic. The tapping has already been done, the publication has already been done. All the damage either side could possibly suffer is all past-tense. What is present-tense is what arguments either side present to justify their actions, and what evidence they are permitted to present in support of their claims.
There has been a discussion on the Linux Kernel Mailing List with regards/dev/random (which I provoked by forwarding a link to a paper critical of it), which GnuPG, OpenSSL and other crypto programs use. Apparently, there are possible attacks which would allow an attacker to calculate earlier random numbers in the sequence, allowing them to guess generated private keys. This seems to be a theoretical attack only and seems to preclude remote exploits, as you'd need access to that device. However, it points to potential risks in the current system.
I'd suggest using LOTS of events to generate entropy (and destroy the Universe) -or- random number generating hardware.
I might even be inclined to be sympathetic...
on
Back to the Moon
·
· Score: 0, Troll
...towards NASA, but they're sponging off the Indian space program, for chrissakes!
THAT is how our great, illustrius NASA is getting to the moon - by outsourcing the R&D and get low-paid foreigners to take all the risks. (Bet you 10:1 that almost all early next-gen manned lunar rockets are built and manned by India, with NASA only using their astronauts when India becomes expendable and the rocket has been constructed.)
I strongly urge all nations with space programs of their own to refuse to cooperate with the American space program unless treated with respect and as equals. I feel reasonably sure that many such programs could reach the moon without much trouble, with no help whatsoever from NASA, before NASA could reach the moon on their own. The more such programs do so, the better. If Russia, China, India and the ESA all get manned mission vehicles that outperform NASA's best at that time, maybe - just maybe - we will see greater cooperation and less nationalism. Total cooperation, total pursuit of technology (and not applause), total openness (none of this "ITAR stops us telling you why our spaceship rammed that satelite" b*shit), and we could see achievements in space taking an order of magnitude less time yet achieving an order of magnitude greater results.
However, the graphics are amazingly crude (think: 1980-1982 quality) and is Windows only. This means that the only people downloading it are (almost by definition) severely mentally disturbed. Fron what I can see from the webpage, the game is designed for a cheap thrill - with such limited content, it is clear that the developer has no real programming skill, no understanding of wargames, no skill at coding AIs... No taste and no sense would seem to be the least of their problems.
A "Columbine" tactical game would certainly be viable. Picture a first-person strategy game where you were a staff member of the school trying to evacuate as many as possible and/or trap and isolate the attackers. That would be an interesting intellectual challenge, offer some fascinating opportunities to examine ways to deal with such situations, etc.
This is merely a very very low-grade imitation of existing kill-em-all games (Postal, and some road-rage game from the UK) and offers nothing that is actually new, interesting or insightful.
First, anyone planning on a high-definition in-flight entertainment system over Bluetooth would have to be nucking futs.
Secondly, if it's used for navigation & engines, it's susceptible to remote hijacks - the Bluetooth "gun" featured on Slashdot before can blast Bluetooth signals over a mile and Bluetooth devices are forever being cracked due to poor security, including poor security of the protocol.
I agree that the cabling in modern planes is excessive and heavy. If we were talking about one optic fibre, that would be one thing, but aviation protocols seem to be point to point, not busses, so you need one physical or virtual connection for EVERY possible combination of end-points. Actually, triple that as they usually use triple redundancy. The aviation protocols are also loosely derived from RS232, so five or so lines are needed for each connection. With triple redundancy, this means that for every given pair of endpoints, you have 15 lines.
Optic fibre is better, but needs repeaters and complicates the endpoints as you have to figure out how to encode the lines into packet form. The problem isn't the data, the problem is the behaviour of the ARINC devices. You've got to mimic the behaviour and characteristice of the hardware they're expecting.
They'd enable multicasting, so that Internet "broadcasts" would consume only a tiny fraction of the bandwidth. (You only need one copy per segment of network, not one copy per receiving user.) They'd also be installing web proxies (such as Squid), caching popular "personally streamed" content on the ISP's servers, etc.
These methods cost nothing to implement, are largely already present and only need to be activated. What we see, instead, is an attempt to triple their already excessive charges, whilst providing a downgraded service. The sad thing is, people are stupid enough that they'll pay more to get less AND think they got a good deal in the process.
They're billing the content provider AND you. They can slow the stream down at either end, so they are implicitly charging twice per packet. (At least. Possibly once each and every hop that the packet traverses.)
They're asking for more from anyone who actually makes use of the bandwidth. What they've sold you is a pipe, but now they want to sell you the right to transmit or receive anything along it. (Since both sides get charged, the ISP now earns three times as much for no extra effort.)
Here is the only good quote from the definitive article (from the movie):
The world's about to end, and here I am, stuck in traffic.
And many great quotes from the TV serial:
That is the dematerializing control. And that, over yonder, is the horizontal hold. Up there is the scanner, those are the doors, that is a chair with a panda on it. Sheer poetry, dear boy. Now please stop bothering me.
Logic, my dear Zoe, merely enables one to be wrong with authority.
Your leader will be angry if you kill me; I'm a genius!
You could augment an earwig to the point where it understood nuclear physics, but it would still be a very stupid thing to do!
Didn't you find two angry men stuck to my car?
What the blazes are you doing in here? Don't you know this area is strictly off-limits to everybody except the tea lady and the Brigadier's personal staff?
"Oh, right now they're far from superior. That's why they left it up to me and me and me."
You may be a doctor, but I am the Doctor. The definite article, one might say.
Deactivating a generator loop without the correct key, it's like repairing a watch with a hammer and chisel. One false move and you'll never know the time again.
The court case was intended to stop Microsoft designing OTHER people's software by means of lawyers and judges. (That's why they refer to "anti-competitive".)
I don't approve of laws designing software, but I have absolutely no problem whatsoever with stopping people abusing laws to prevent software from being designed. I also have no problem with laws that enforce progress.
(The State of Oregon recently received some thinly-veiled threats from Microsoft's CEO over Oregon's active support for Open Source - both towards Oregon and towards all Microsoft shops in Oregon. Although not a part of the DoJ lawsuit, and therefore probably not a part of this review, I would feel a lot more comfortable if States receiving such threats reviewed their legality. Last I heard, "demands with menaces" was not considered an OK activity.)
My uncle is a Guernseyman, my father & grandfather taught at Elizabeth College, and I have visited there (and Herm) many times. So although not a native of Guernsey, I feel I know it as well as any non-native could.
A word of warning - if you have a metal detector, be very careful near the beaches. Back in the late 70s, I found an unexploded German bomb about three inches below the surface of an unsurfaced beach car park. (I reported it to the police - after carrying it to the house I was staying in, as I wanted to wash it off to get a better look. Ah! Those were the days!)
That part is definitely wrong. The climate - and the weather - are 100% deterministic systems. There is nothing "random" about the climate, it follows very exact mathematical and physical laws, with no exceptions. Now, your geography teacher may have meant "non-predictable", and that would be true. The climate is a chaotic system, which means it is very sensitive to initial conditions, has no stable points and the mathematical system representing it has no differential.
(The difference is that a random system has nothing predictable about it. It can do absolutely anything at any time, merely following the probability distribution for that system. However, the exact state at time T can be known - it merely doesn't mean a whole lot. A chaotic system, on the other hand, has definable patterns, definable mechanics and definable structure, but you can NEVER know the exact state for ANY time and small differences CAN - but won't always - cascade into large changes.)
If you were to look at the climate as far back as we can reliably know it, you will see oscillations between ice ages and warmer periods. The troughs and peaks appear fairly random, but really they aren't. The climate can be approximated (badly) as a simple oscillating function, but that's pretty crude. Actually, there's a greater correspondence between 10 years stock prices for wheat and 100,000 years of global temperature than there is between climate and a sine wave. (See: "Fractal Geometry of Nature", Mandelbrot, B., for more details, as the margins here are much too big - err, small.)
I would also be willing to bet that the change in climate as a function of the change in climate composition is also very deterministic. As we're talking systems that appear to be oscillating, my best guess is that the ratio of the peak intervals of different types of oscillation with small differences in climate composition will always be Feigenbaum's Constant, as that's usually the case in chaotically-produced pseudo-oscillations.
Now, despite all this talk of chaos, lack of absolutes and so on, the climate is very predictable in general form. No great surprise there - if you generate the Mandelbrot Set, or the Lorenz owl-mask, you expect to see the same general shape each time. That is not going to change. The same is true with the climate... for now. The climate is orbiting a bunch of Strange Attractors, as per the Lorenz owl-mask. We know the general shape and we know the general effects of altering the various parameters.
There is a problem, however. If the climate were to jump from the current set of Strange Attractors to any other set, the climate would change relatively rapidly and definitely counter to any model that relies on the current patterns holding true.
What could cause such a jump? When could it occur? Well, that's the problem. Strange Attractors are not like nice, neat gravitational sources, you can't see them, and they have no physical existence, they are merely a product of irreducable mathematical problems. They could, however, cause the planet to boil or freeze the moment the system strays too far. (If you don't know which Strange Attractor the climate would switch to, you cannot make any useful prediction from past trends.)
For those wandering internationally, here are the following viral fashions currently in vogue:
England: Not sure, but it was either designed by HRH The Prince of Wales or won an award from him for innovation
The Channel Islands: A "Stone De Croze" outfit (it's important to be original)
Australia: A cross between "Crocodile Dundee" and Rolf Harris
Germany: Doesn't matter - the Chaos Computer Club will upload Blinkenlights and use the entire street for a gigantic game of Pacman
Denmark: Is there such a thing as a viking mermaid?
America: Masks of any kind will automatically transform into the face of one of the FBI's ten most wanted, at random, and the jacket will carry an amazing image of an AK-47 that you'd swear was real, with a net impact on Halloween candy sales
Absolutely any abandoned rock quarry: Random assorted costumes from BBC science fiction series
I agree 100% with the parent post. Some suggestions for computers and why:
Sinclair ZX80 - Start of European craze for personal computers
Sinclair QL - An 8-bit computer that emulated a 32-bit computer, hated it at the time but it was significant
PET 2001 - Built-in tape deck & monitor!
PET 8032, but you'd want the 256K RAM pack and possibly the PIC chip (which gave it basic graphics capability)
BBC Micro (either model A or B) - Massive I/O capacity
Acorn Archimedes - Early GUI, early RISC home computer
Transputer (any model) - Entire 32-bit parallel computer on a chip & usu. on eBay. Most times was stand-alone or as a processor board for a PC, but also came as a plug-in module for the Amiga to boost some games
Sun Sparcstation I - Computer in the monitor
Osborne 1 - First portable computer
Not sure who made the All-In-One (computer, modem, printer, keyboard and monitor in one semicircular unit)
And my challenge to you... Most of these computers had at least a serial port. I don't believe the ZX80 did, though it may have had a tape port. Find a way of building a gigantic network using these random archaic technologies. The difficulty in building something that can convert between the myriad of formats and speeds would be a nightmare. So why do it? Partly to get a better appreciation for the internals, partly because many who were familiar with the machines would tell you it's impossible, but mostly because running a gigantically networked fishtank program across an impossibly diverse set of machines never built for parallel activity would completely f with the minds of computer historians.
(This idea is not unlike the person who, for the 50th anniversary of the Manchester Mk. 1, wrote a program to turn the status registers, which were displayed, into a scrolling banner. Totally sick and inappropriate usage of such a machine - so much so it was brilliant!)
It was closer to flesh-eating bacteria. Piracy, like the poor, will always exist. There are ways to limit the scope, though. In the case of the RIAA, hiring fewer prostitutes and spending less on cocaine would probably be an excellent start. The savings should be enough to maintain the profit margins even after slashing CD prices in half.
Slashdot Mirror
These are all cases where the truth should, very reasonably, be of limited availability. They are, I acknowledge, fairly extreme examples but that is because for the most part it really takes an extreme case to justify such an extreme measure. (As you note, blackmail is also such a case - I guess it's still extreme for the individual concerned, though.) In my examples, I also note the harm involved, so I guess the second constraint is when intended harm exceeds good. All of the parent's examples and mine also concern themselves with speech that is not "freeing" or "unprejudiced", but is for the express purpose of "control" and "manipulation" - either of the target audience and/or, where different, the targets of the speech.
Where a reasonable person can conclude that the speech is within such an extreme circumstance that the consequences will be disproportionate to the speech itself, AND where that speech is intended to cause disproportionate harm, AND where that speech is not truly "free" but laced with spin, THEN I can see no sensible objection to constraint on that speech. In fact, it better well had be constrained*.
Where a reasonable person can conclude that the speech is in a moderate or benign contect, AND that the speech is intended to bring about good, AND where the speech is not only truthful but devoid of spin or propoganda, THEN that speech should be protected utterly and unconditionally.
This leaves all other permutations. I would argue that as you approach the absolute of the extreme case, you will get fewer and fewer cases that are clear-cut cases of where speech should be unconditionally free. However, I believe absolutely that the assumption should be one of speech being free and that those who would limit it should prove unconditionally and beyond all reasonable doubt that the listed factors are present and that they outweigh all other factors.
*Where a person or an agency deliberately and knowingly sets themselves up in a situation where exposure would seriously discredit that person or agency - particularly if it is a crime, a judge should have the power to rule that they are not entitled to legal protections from the consequences of their own actions.
Under NO circumstances should a Government be able to tell a court what should be secret and what shouldn't. I do not believe in such a right and see no justification for it, if only for the simple reason that that violates the separation of powers and also creates a conflict of interest.
I absolutely love playing with new technology - can't get enough arcane, bizare and downright weird programs that do stuff that's novel or just plain strange. I hope ACME fits into this category, but as the above list shows, it has tough cometition before it qualifies as new & interesting (at least to me). Being able to store scriptlets in one window to apply to another might qualify, if there's some new tangent to it. Oh, and I'd have to be sure that the method used to apply scripts in this way did not pose a security issue -- the vast majority of all the viruses currently for Windows are macro viruses, and the early (AT&T) history of Unix includes tales of viral backdoors.
Trust me, I want to be convinced, if for no other reason than I'm running out of new programs to play with. The nightmare of withdrawl symptoms, suffering from stale sameness... It doesn't bear thinking about!
KROC is here. The other is some long-forgotten package, as opposed to a long-forgotten language.
I don't know if KROC has been ported to Plan9/Inferno yet, but it damn well should be.
The backbone providers are unlikely to care that much - it impacts a little business, but most make money off their inter-corporate and inter-Governmental lines. The more the Internet degrades, the more high-priced services the major vendors can sell and the more copper/fiber the telecos can charge for. I don't see much of a motive to fix things here.
The vendors further up the chain don't need to care much, either. The companies on the Internet can't gain by switching ISP, because it's the backbone that's broken and they'll have to go through it to reach the peasents - err, home users anyway. The corporations that sell over the Internet don't lose any sales, as a person who is going to buy from an online store is likely to be doing other stuff and won't go out to the stores, so they'll be back. Home users, for the most part, are ignorant enough to think AOL and MSN are really neat ideas, have no clue what the Internet involves, what needs fixing or why, and is likely to pass it off as someone else's problem anyway. And those who ARE smart enough are Libertarian enough that they won't Unionize and DEMAND the fixes that damn well should be made.
(IT users and IT professionals should stop with the "unions are evil" crap - no organization is any more evil than the people in it - and collectively insist that the defects be fixed. No ifs, no buts, no maybes, no excuses, no delays - these kinds of attacks SHOULD be impossible and COULD - very cheaply - be made impossible. But nobody is going to even take the cheap option without a fight, if there's an even cheaper option of apathy open to them.)
Given that corporations have been very careful to put their primary servers inside the corporate firewall to protect them, blasting the system load and application status over the public internet would seem to be retrograde. If there's a "Big Brother" or MRTG server in the DMZ that can see the internal servers, then there's a machine crackers can use to bypass all internal security measures. And what are these Indian IT guys going to use to log in, anyway? There's a frightening possibility that connections will be through raw telnet or the most basic setting on Terminal Services, and that corporations will simply plain-text e-mail the usernames and passwords required to the Indian company managing their systems.
All this ignores the bandwidth issue. Most companies can barely afford to get the pipes they need. If you're now going to fire RRD databases and SNMP streams over to India, you're going to eat into that bandwidth. As soon as you throw in virtual consoles for GUI management (and that's the most likely way they'd control most of the boxes), you're talking severe network hogs. This means that corporations with insufficient bandwidth will kill their connections entirely or have to spend more on additional bandwidth than they're saving by outsourcing.
In the same way that a trade secret that becomes public ceases to be protectable as a trade secret, I would have though that this would cease to merit any protections as it is self-evidently no longer secret, whatever the state may say.
So, on the basis that state secrets does NOT appear to be a valid piece of Common Law, and that there is no secret left to protect, I can see no justification for quashing this evidence. Furthermore, as the documents HAVE been published openly, AT&T have lost all rights to their claim of trade secrets, and so I can see no obvious justification of the evidence even being sealed. We already know what the bulk of it says, as it's online!
The argument over who is right and who is wrong is, in this case, largely academic. The tapping has already been done, the publication has already been done. All the damage either side could possibly suffer is all past-tense. What is present-tense is what arguments either side present to justify their actions, and what evidence they are permitted to present in support of their claims.
I'd suggest using LOTS of events to generate entropy (and destroy the Universe) -or- random number generating hardware.
THAT is how our great, illustrius NASA is getting to the moon - by outsourcing the R&D and get low-paid foreigners to take all the risks. (Bet you 10:1 that almost all early next-gen manned lunar rockets are built and manned by India, with NASA only using their astronauts when India becomes expendable and the rocket has been constructed.)
I strongly urge all nations with space programs of their own to refuse to cooperate with the American space program unless treated with respect and as equals. I feel reasonably sure that many such programs could reach the moon without much trouble, with no help whatsoever from NASA, before NASA could reach the moon on their own. The more such programs do so, the better. If Russia, China, India and the ESA all get manned mission vehicles that outperform NASA's best at that time, maybe - just maybe - we will see greater cooperation and less nationalism. Total cooperation, total pursuit of technology (and not applause), total openness (none of this "ITAR stops us telling you why our spaceship rammed that satelite" b*shit), and we could see achievements in space taking an order of magnitude less time yet achieving an order of magnitude greater results.
A "Columbine" tactical game would certainly be viable. Picture a first-person strategy game where you were a staff member of the school trying to evacuate as many as possible and/or trap and isolate the attackers. That would be an interesting intellectual challenge, offer some fascinating opportunities to examine ways to deal with such situations, etc.
This is merely a very very low-grade imitation of existing kill-em-all games (Postal, and some road-rage game from the UK) and offers nothing that is actually new, interesting or insightful.
Secondly, if it's used for navigation & engines, it's susceptible to remote hijacks - the Bluetooth "gun" featured on Slashdot before can blast Bluetooth signals over a mile and Bluetooth devices are forever being cracked due to poor security, including poor security of the protocol.
I agree that the cabling in modern planes is excessive and heavy. If we were talking about one optic fibre, that would be one thing, but aviation protocols seem to be point to point, not busses, so you need one physical or virtual connection for EVERY possible combination of end-points. Actually, triple that as they usually use triple redundancy. The aviation protocols are also loosely derived from RS232, so five or so lines are needed for each connection. With triple redundancy, this means that for every given pair of endpoints, you have 15 lines.
Optic fibre is better, but needs repeaters and complicates the endpoints as you have to figure out how to encode the lines into packet form. The problem isn't the data, the problem is the behaviour of the ARINC devices. You've got to mimic the behaviour and characteristice of the hardware they're expecting.
"broadcasts" would consume only a tiny fraction
of the bandwidth. (You only need one copy per
segment of network, not one copy per receiving
user.) They'd also be installing web proxies
(such as Squid), caching popular "personally
streamed" content on the ISP's servers, etc.
These methods cost nothing to implement, are
largely already present and only need to be activated. What we see, instead, is an attempt to triple their already excessive charges, whilst providing a downgraded service. The sad thing is, people are stupid enough that they'll pay more to get less AND think they got a good deal in the process.
They're billing the content provider AND you. They
can slow the stream down at either end, so they
are implicitly charging twice per packet. (At
least. Possibly once each and every hop that the
packet traverses.)
They're asking for more from anyone who actually
makes use of the bandwidth. What they've sold you
is a pipe, but now they want to sell you the right
to transmit or receive anything along it. (Since
both sides get charged, the ISP now earns three
times as much for no extra effort.)
And many great quotes from the TV serial:
I don't approve of laws designing software, but I have absolutely no problem whatsoever with stopping people abusing laws to prevent software from being designed. I also have no problem with laws that enforce progress.
(The State of Oregon recently received some thinly-veiled threats from Microsoft's CEO over Oregon's active support for Open Source - both towards Oregon and towards all Microsoft shops in Oregon. Although not a part of the DoJ lawsuit, and therefore probably not a part of this review, I would feel a lot more comfortable if States receiving such threats reviewed their legality. Last I heard, "demands with menaces" was not considered an OK activity.)
A word of warning - if you have a metal detector, be very careful near the beaches. Back in the late 70s, I found an unexploded German bomb about three inches below the surface of an unsurfaced beach car park. (I reported it to the police - after carrying it to the house I was staying in, as I wanted to wash it off to get a better look. Ah! Those were the days!)
(The difference is that a random system has nothing predictable about it. It can do absolutely anything at any time, merely following the probability distribution for that system. However, the exact state at time T can be known - it merely doesn't mean a whole lot. A chaotic system, on the other hand, has definable patterns, definable mechanics and definable structure, but you can NEVER know the exact state for ANY time and small differences CAN - but won't always - cascade into large changes.)
If you were to look at the climate as far back as we can reliably know it, you will see oscillations between ice ages and warmer periods. The troughs and peaks appear fairly random, but really they aren't. The climate can be approximated (badly) as a simple oscillating function, but that's pretty crude. Actually, there's a greater correspondence between 10 years stock prices for wheat and 100,000 years of global temperature than there is between climate and a sine wave. (See: "Fractal Geometry of Nature", Mandelbrot, B., for more details, as the margins here are much too big - err, small.)
I would also be willing to bet that the change in climate as a function of the change in climate composition is also very deterministic. As we're talking systems that appear to be oscillating, my best guess is that the ratio of the peak intervals of different types of oscillation with small differences in climate composition will always be Feigenbaum's Constant, as that's usually the case in chaotically-produced pseudo-oscillations.
Now, despite all this talk of chaos, lack of absolutes and so on, the climate is very predictable in general form. No great surprise there - if you generate the Mandelbrot Set, or the Lorenz owl-mask, you expect to see the same general shape each time. That is not going to change. The same is true with the climate... for now. The climate is orbiting a bunch of Strange Attractors, as per the Lorenz owl-mask. We know the general shape and we know the general effects of altering the various parameters.
There is a problem, however. If the climate were to jump from the current set of Strange Attractors to any other set, the climate would change relatively rapidly and definitely counter to any model that relies on the current patterns holding true.
What could cause such a jump? When could it occur? Well, that's the problem. Strange Attractors are not like nice, neat gravitational sources, you can't see them, and they have no physical existence, they are merely a product of irreducable mathematical problems. They could, however, cause the planet to boil or freeze the moment the system strays too far. (If you don't know which Strange Attractor the climate would switch to, you cannot make any useful prediction from past trends.)
And my challenge to you... Most of these computers had at least a serial port. I don't believe the ZX80 did, though it may have had a tape port. Find a way of building a gigantic network using these random archaic technologies. The difficulty in building something that can convert between the myriad of formats and speeds would be a nightmare. So why do it? Partly to get a better appreciation for the internals, partly because many who were familiar with the machines would tell you it's impossible, but mostly because running a gigantically networked fishtank program across an impossibly diverse set of machines never built for parallel activity would completely f with the minds of computer historians.
(This idea is not unlike the person who, for the 50th anniversary of the Manchester Mk. 1, wrote a program to turn the status registers, which were displayed, into a scrolling banner. Totally sick and inappropriate usage of such a machine - so much so it was brilliant!)
Nobody would believe North Dakota was Mars. North Dakota is far colder and far less hospitable.
It was closer to flesh-eating bacteria. Piracy, like the poor, will always exist. There are ways to limit the scope, though. In the case of the RIAA, hiring fewer prostitutes and spending less on cocaine would probably be an excellent start. The savings should be enough to maintain the profit margins even after slashing CD prices in half.
Everyone. And if the illegal trade in body parts in the US is as big as made out, we already have the rest of the movie's elements.
All we need now is for people to swell up and explode, and for TV execs to suppress truth. 3.5 seconds into the future...
Lawyers would be more of a deterrent than animals, but it is unclear whether the radiation would kill them or simply enhance their undead superpowers.