You can set a soap opera anywhere. You can set a soap opera in a few rooms, costing next to nothing, and get exactly the same content. In consequence, most studios are going to opt to pay $2 per episode over $2 million. If you want something set in space (which means the cost is going to be high), you need content that justifies that cost. That means the Universe it is set in needs to be crucial - otherwise, what's the point in having it there? It also means the science has to grip people enough that the audience feels as though it is there.
As Tolkein once noted in his essay "On Fairy Stories", if you have to suspend disbelief, the author failed. The true test of a good story, Tolkein argued, is that you shouldn't have to suspend anything. It should feel real to you, at least to the point where things "just make sense". Lengthy technobabble and explanations shouldn't be necessary.
Again, though, most of this is equally true, no matter what the setting. When the setting is exceptionally difficult, expensive or unusual, therefore, the setting has to be relevent. Asimov's Foundation series could not work in a Wild West setting. The original Star Wars trilogy uses space to express enormity - something no other setting could provide - so whilst the story is a fairly trivial quest in and of itself, it couldn't carry the range of expression in any other setting. The Matrix is also a fairly trivial quest, but requires the deformable, maliable nature of virtual reality in order to cover the metaphysics of the story. It simply couldn't work anywhere else.
Serenity, on the other hand, had nothing new in it and required nothing from the setting. The story looked like a simple mesh of a cowboys-and-indians western with Matrix-style combat, Libertarian vs. The Evil Commies politics, and stole the "alien enemy" from an Asimov short story.
Generally, when people have ideas, it is not in isolation. They combine ideas in their mind and produce a result, they don't just pull a fully-formed idea out of thin air. The ideas are mixed in a way not a million miles from how genes are mixed in living organisms.
If you want to regard mental processes as a form of sexual reproduction in the most abstract form possible, then I guess it would describe the essential underlying mechanisms involved in how people think, and would certainly go along with a fairly common attitude amongst engineering types that the product of their mind's work is their "baby".
...they're stupid? Well, maybe that's a little unfair - many e-mail clients don't support PGP or GPG. However, Thunderbird DOES support X.509 certificates and therefore they could certainly use X.509 to sign their e-mails. I believe X.509 is also the system used by Outlook (bah! bumbug!) and other "popular" e-mail clients. Dunno why - there are more people with GPG keys than X.509 certificates, but that's what's supported at present. At the very least, signature support DOES exist and COULD be used, so damn-well SHOULD be used by E-Bay, PayPal, Amazon, banks - the usual targets of phishing scams.
Actually, even if users didn't check, it would be easy enough for someone like E-Bay to run a promotion targetting ISPs - "if you add our automatic phish frier, you can carry our flashy 'secure e-commerce' logo and be listed as a 'trusted' partner!", where the phish frier simply drops any e-mail that has a FROM claiming to be from E-Bay but where the signature is missing or is incorrect.
ISPs love anything that makes them look good - especially when it doesn't cost them anything. All you'd really need is for a couple of the major players to provide ISPs with such filters, and for just one or two of the major providers (AOL, Comcast, etc) or major webmail providers (Yahoo, Hotmail, Gmail) to install them. I would see that as very doable and would cut down on the phishing scams, even if they didn't eliminate them.
PGP/GPG signatures - The software exists but would preclude traditional webmail. The only way it would work on webmail is if your machine has a cerification server the webmail could send the e-mail to. The cerification server would then digitally sign the e-mail and return it to the webmail server. Very very few regular clients support PGP or GPG. The only way to make this mobile is to have the encryption keys stored on a USB device and even then, not all libraries or cyber-cafes allow you to plug in USB devices.
X.509 signatures - A few more e-mail clients support X.509 than support PGP or GPG, but you'd still need the trick with the relay to digitally sign webmail. It's trickier to make mobile, as those programs that do exist generally assume a central, on-machine store for X.509 certificates.
X.400 e-mail - This specifies all kinds of authentication and verification. If anyone actually implemented it on a modern OS, it would be awesome. It would also be very, very heavyweight, hard to maintain and totally non-standard. It is also massively encumbered, so the only way to implement it to F/OSS standards would be in a country that doesn't respect Intellectual Property.
Intermediate, tamper-proof signing - This is not as strong, but it is something. The idea would be for all (E)SMTP servers to not just attach a header saying they sent the message forwards, but would add an X- header that digitally signed the message as OK. The server would also need to check that e-mails they received were indeed signed and that the signature matched both the claimed last sender and the message. This would eliminate phishing attacks involving fake mail servers and would have the advantage that the user need not do anything (including update anything), but would have the disadvantage that it would only authenticate servers, NOT users.
PGP/GPG use RSA or ElGammel public-key encryption to store a secret key, which is then used to decrypt the actual message. This is faster than using RSA or ElGammel for the entire message, but is only as strong as the weakest algorithm in the chain. There are substantially faster public-key algorithms, but they are either known to be broken (HFE), heavily encumbered (HFE, NTRU) or regarded with suspicion (ECC).
The reason this is important is that no authentication scheme can be any better than the method used to prove or validate the identity. (Duh!) Therefore, it is essential that the authentication scheme can be trusted to do what it says - authenticate that the message originated with the person the message claims to be from, with absolutely no possibility of the message originating with anyone else or being modified en-route (except for relay headers).
This means that PGP-style encryption does not prove identity. The message must be signed to prove identity, if the public key is only used to hide a secret key. Even then, with hashing algorithms tumbling like dominos (with or without sparrows helping), you need to be somewhat strict about what method is used for authentication. In four or five years, it is entirely possible to imagine skript kiddies being able to fake MD5-based signatures and for organized crime syndicates to be able to fake SHA1-based signatures. As these are the two largest sources of phishing scams, any approach which they are likely to completely defeat within the lifetime of any standard adopted is useless.
An Internet-based protocol can be considered of having a life-expectancy of 20 years, with no substantial modifications being possible. Maybe 30 years, in some cases, even if the infrastructure is incapable of handling the load. (IPv6 demonstrates that.) That does not mean any given e-mail must be proof against 30 years of concentrated effort. Digital certificates and encryption keys usually expire after a few months or a year, so we don't care if anyone breaks a particular key after that. The ideal signature scheme, then, has only to be reasonably secure for the same few
Well, for most places, anyway. It is just possible for the Queen to declare England to be in a state of Constitutional Emergency, in which case she can usurp power from the Government and lock Tony Blair in the Tower of London. Chances are, she won't, though.
In Britain, that leaves the House of Lords, who can threaten to veto all legislation from now to eternity. Not a whole lot anyone can do to stop them, either, as they're not elected and it's next to impossible for anyone to impeach them. Still, Tony Blair has threatened to eliminate the entire House (and one can never be quite sure on the interpretation on that) if they defy him too much.
There's also the European Court of Human Rights, but they have only limited powers to enforce their rulings. Generally, though, Governments will bow down to them. At least, in public. The European Court is interesting in all of this, because they are very unpredictable.
The last great hope of Europe is, ironically enough, the British Intelligence network. "Huh?" you might very well say. They have apparently brought down previous British governments (such as that of Harold Wilson), when they have believed the Government had become a danger to Britain. Not a million miles from the way the Watergate scandal broke in the US, only there is evidence that MI5 collectively acted against their superiors in Whitehall, rather than a single individual acting on their own.
Is it even remotely likely MI5 would bring down Tony Blair? Probably not. Would it be even possible? Oh, certainly. And, as noted, there is some basis for believing they've taken matters into their own hands before.
But this is mostly by-the-by. George Bush can't be re-elected and his supporters control all divisions within the Government, so he has no consequences to consider. In Europe, politicians are forever looking over their shoulders. They are always at risk from scandals or other PR disasters. They are therefore far more likely to listen to people, provided it is done right and they consider the risks too great. That is why an organized voice of the IT community must present a coherent case to them. They will listen - they can't afford not to.
Those in London had rather more to worry about, when those cameras went up. The IRA were literally driving trucks with thousand pound bombs in the back, which made a lot of people very nervous. Things have settled down somewhat since then.
It is also important to remember that although data retention laws require information to be available for security reasons, the Data Protection Act prohibits making that information available to anyone else. In fact, most of Europe has incredibly strict privacy laws - along with laws prohibiting the trade of such information to organizations and companies that are outside the jurisdiction of those privacy laws.
A good solid campaign by European technophiles, to remind Euro MPs (and regular MPs) about the British and European privacy laws with regards to personal data, especially when coupled with reminders of Microsoft's extremely dodgy past on security issues, would be likely to derail Microsoft's efforts entirely, as their proposals are technically illegal and politicians in Europe - at least for now - are eager to NOT be seen aiding and abetting lawbreakers. In fact, a solid-enough campaign that also brought in Microsoft's status in Europe as a monopolist guilty of breaking trade laws might potentially finish off Microsoft in Europe entirely.
Before anyone marks me as rabidly anti-Microsoft (I'm not rabid, I just don't like them), this post is not a condemnation of the corporation. It is merely pointing out that their proposal violates EU privacy directives and assorted national laws. This is more likely ignorance than malice, as America has no concept of privacy, but that simply isn't relevent. It would be relatively easy for someone to spin this in a way that would leave politicians with very cold feet. Politicians aren't generally brave - that's not how to get re-elected. Politicians are professional cowards. No sane coward is going to want to be seen breaking the law - or even potentially breaking it - right now. If that were to be how opponents presented it, most politicos would back off very very quickly.
Well, it is certainly true that a book I have from 1750 describes thunder as the explosions of evaporated gunpowder, so a lack of understanding is certainly possible. On the other hand, we're in the early 21st century now, so I would have expected somebody to have figured this all out in the intervening hundred to hundred and fifty years.
...the regulation has to be independent. Government regulation doesn't work, in general, because Governments are frequently corrupt and corrupt regulation simply increases positive feedback. However, to be effective, the regulators need the powers of a Government.
If there was a "simple" solution to this problem, it would have been solved by now. Nonetheless, I do believe that the problem is solvable and that the solution does involve regulation mechanisms. I suspect that the solution must involve some sort of "blind" element (a jury pool system, for example) so that it is not possible for any particular side to know who to bribe. It must also have people with the necessary knowledge but who are unlikely to have been drawn into the system so far as to have become corrupt themselves. The only way you'd get that combination is to introduce a mix of recent graduates and the self-taught. The remainder of the regulatory body should be directly elected as per any other representative, only using a proportional representation system to prevent domination.
It still wouldn't be perfect, but the randomness of one component, the youth of another and the capricious nature of PR elections would make most efforts at corruption extremely difficult and VERY expensive, and therefore probably much reduced.
The lights back then probably were not cars - unless they were Delorians... However, other posters have mentioned that there are other requirements - high humidity being one. My guess is that the distant highway in question would also need to have relatively low humidity. Furthermore, I would guess that if you were to draw a diagram, showing the observer, the apparent position of the lights, the boundary between the two air masses, and the cars, you'd find that the light is being bent by the amount you would expect from the difference in refractive index.
Now, how does this relate to the lights in the 1800s? Oh, quite easily. I suspect the lights were quite probably fires, but considerably further away and in a completely different place than the observers had expected - which is why they never found anything.
As for people chasing the lights and never reaching them (according to another poster), this is exactly what you expect from an optical illusion from refracted light. Most people have seen this with rainbows, which are also caused by refraction through water droplets. It's the same mechanism, so you get the same "moving" effect. Duh.
In fact, once people had observed they could not "approach" the lights, the physics of it should have been obvious. There aren't many types of illusion which work that way. You can approach a mirage, for example, but it vanishes when you get "too close". If you shine a bright light onto fog, you will get reflected light from it. Etc.
The law of supply and demand requires that demand will always rise to the point that supply is incapable of meeting that demand to such a degree that prices rise to the maximum that the market will bear at that time. If supply exceeds that value, you will get undercutting, which will lead to an artificial inflation of demand until undercutting is no longer feasible.
In other words, a shortfall is not a product of a particular society or of outsourcing, it is a simple product of economic forces. Or, at least, those forces that lead to stable economies. Most economies are highly unstable, which means that instead of reaching a dynamic equilibrium, they are highly chaotic. Indeed, much of what is known about chaos theory comes from people studying economics, so there have been fringe benefits.
The instability largely comes from political corruption, monopolies of any kind, insufficient worker protections, price gouging, accounting fraud and other such elements, where the normal feedback mechanisms cannot work or are even actively prevented from working. What you end up with is a positive feedback loop, which leads to the whole boom/bust system that America in particular is infamous for, and which India is likely to lurch into, because it has absolutely no comprehension of the kinds of controls you need to keep the system sane.
The problem with America is that Government regulators are largely financed by the corporations they regulate. This is like having a computer virus scanner asking a virus if it is present. You think it's ever going to tell the truth?
India is in a worse mess, because bribes and back-handers are common practice on the sub-continent. That's like installing a computer virus AS the virus scanner. And sooner or later, jobs will migrate to ever more opaque, ever less accountable, ever more corrupt systems, because those are cheap - in the short-term, at least. Corruption is always expensive in the long-term, just not necessarily for the ones guilty of it.
The only way for India to build a stable economy out of the outsourcing is to have the Government conduct strict, well-defined oversight in a 100% independent fashion. However, that will drive up costs considerably. It will also drive up standards, but accountants don't include standards on the balance sheet and it's the balance sheet that matters. Stability in India will, therefore, effectively end outsourcing there. Which means that India must choose between being popular or being viable. It can't be both.
It means we're opposed to those who are opposed to the opposition of those who oppose the opposition's overlords. I guess that makes it The Story of "O".
I would have agreed with you. Then came the "headbanger" virus, which smashed disk drive read heads against the end buffers which would misalign them or break them entirely. X11R4 configuration files were notoriously dangerous, as you plugged in timings directly - ever heard a monitor scream? I have. CPUburn (it's listed on Freshmeat, so no excuse for not knowing about it) puts the CPU into an exceptionally tight loop, using instructions specifically selected to generate heat. If the CPU is not properly cooled, it can actually destroy the CPU.
Let's see, what other examples are there? Incorrect BIOS writes, as another poster noted, are Bad News, but they are technically recoverable. Well, usually, anyway. If there is nothing to reset the BIOS to, it can be much more troublesome. Hard disk thrashing is a bigger problem, as that can cause an actual head crash.
(So that non-geeks can get some idea of why head crashes are BAD, your typical hard drive has a diameter of 3.8" - and therefore a cicumference of about 12". It revolves 7200 times a minute, which means the outside of the disk is travelling at a speed of 1432.6" per second. Now, imagine an extremely delicate read-head striking that disk. Now picture that scene from Star Trek:TNG where the saucer section from the Enterprise makes an emergency crash landing. That is what your hard drive now looks like.)
Other ways software can cause hardware damage: Loading microcode into the CPU that does the same as CPUburn, only a great deal more savagely. Try googling for Service Pack 6 and you'll find machines that could not be rebooted. At all. They were utterly dead. Buggy drivers on buggy hardware aren't a good combination. If you've an old BBC Micro handy, try the following instruction:
10 DIM a%(1,1,1,1,1,1,1,1,1,1,1,1,1,1)
The Commodore PET allowed you to alter video timings, but was never designed for such modifications. It was therefore possible to cause permanent damage to the display.
These are only a few examples, but computer history is littered with them. The Manchester Mk 1 had far more opcodes than instructions, with unused opcodes running through utterly unpredictable paths in hardware. I think it used 32-bit addresses, but it only had 32 words of memory. Again, going outside of physical memory was unpredictable. You programmed that machine with EXTREME care.
Software, these days, is designed around workflows and dataflows. In other words, software is no longer simply a device in a designated state onto which data is mapped. It has become something that has a time element intrinsically associated with it. (If the software had no "concept" - for want of a better term - of time, it could not model a flow. A flow requires a past, a present and a future.)
However, it is highly inefficient to modify data in a purely serial manner. Serial went out with punch tape. You are always going to have to slice data in some form or other, because data will typically have far more dimensions to it than can be entered or displayed.
The problem is, most slicing mechanisms work on instantaneous snapshots. They are "horizontal" slices, if you like, across a moment in time. But flow-based programs don't just have an instantaneous state and therefore don't have any reason to be limited to just receiving data as if that was all the program could handle. Flow-based programs are built around the system/model as a whole, over all possible paths and eventualities. Feeding data into them a single time-slice or event-slice at a time is no more efficient than writing a program to plot a sine wave by manually entering fixed values at fixed points.
Let us take this one step further. Not only have people broken away from the instantaneous state, they have broken away from sequential ordering. Programmers, for the past twenty to twenty-five years have been much more interested in multi-threaded, multi-data, anticipatory, event-driven programs and computer architectures.
The only "sane" way to handle the I/O for such software is by allowing time to be an element in the display. And this is fine. Humans are great at recognizing patterns - it is what our brains are optimized for. And there are no patterns in a single slice, because by taking a single slice, there is nothing to relate to and therefore no patterns to observe.
The only interest manufacturers and developers have had in 3D displays, though, is allowing people to display a more "complete" or sophisticated slice. They're still completely missing the point that interactive computing isn't batch processing on tranquilizers. If you're going to have interactive computing, make use of having a human at the console. Give the human patterns! Allow them to manipulate those patterns directly. In other words, allow them to directly manipulate the change of the data, not just the value of the data.
If you're going to have "complete" slices via a 3D display, then ergo, you must need a 4D display to manipulate the way the data changes, particularly if you want to be able to see how the flows split and recombine, as a single slice cannot capture information about the path(s) you are NOT on. Furthermore, to go back to the sine example, slices are only good for discrete events, they're not useful for continuous systems. However, 99.9% of reality is continuous. (Which is why computer models are generally crap. Discrete approximations of continuous systems are almost never going to be any good.)
High-performance computing has reached the limits of sliced-and-diced discrete-event I/O formats. If HPC is to progress much further in human-computer interfaces, they will have to display change as something more than just frames in an animation. It isn't realistic to the user, isn't logical to the program and isn't producing useful results anyway. Since it is generally the HPC labs that have the Really Big Money, this is where you will see any new display technology.
LCD monitors that can use polarized light to display full-color 3D images - hey, that's cool. The systems at SC|05 were most impressive for what they were. But they're not system images, they're still snapshots. People want to manipulate systems and to do that they will need to "see" systems and interact directly with them. And 3D just isn't enough. You need 4D for that, whatever 4D might actually mean in a p
Those kinds of superstitions (avoid all products that have a version number that ends in.0) are based on the premise that certain releases are more likely to be rushed than others, so are more likely to have bugs. In reality, any release is as likely to be rushed as any other, any product that is rushed has a raised probability of having defects, and any defect - however small - can (in principle) be totally lethal to a computer program.
If you look at Microsoft's scorecard for service packs, sure, you can find even-numbered ones that locked systems solid. You can find odd-numbered ones that did that, too. The only method I can suggest for safely installing a service pack is to wait until someone publishes a regression test on that SP on the machine you use with the hardware you have. The next-best is to find a combination of tests which cover the arrangement, but that's not as good, as interactions between components can be critical.
The third option is to make extensive backups, then try installing the service pack. If it doesn't work, restore from backup. The problem with this is that there are rare bugs that cause actual hardware damage. I only recall reports of such damage for one Windows service pack, but I am definitely unhappy about the fact that there are any such reports at all. It is for that reason I relegate this to being a third option, because (IMHO) it is preferable to let someone else fry their computer - doubly so as Microsoft won't take responsibility and manufacturer warranties are usually voided if you change the system at all (such as installing service packs).
...that claimed that 10% of homeless people there had a degree or better, so it's certainly possible a decent percentage are former programmers. Definitely, I have talked to homeless in Manchester who were former network engineers and mainframe engineers, so it does happen.
On the other hand, I know that many early machine developers at Manchester University, where the first stored-program computer was built, remained on the faculty until they retired. I also know people from Imperial Computers Limited (a UK mainframe company) who were extremely well-off.
The serious answer, then, is that it varies.
As for some of the not-so-serious answers others have offered... well, I always was suspicious of University caffeteria food...
I knew there had to be someone out there who liked KGI (Linux' kernel graphics infrastructure, which incidently did allow you to use X with kernel-level accelerated drivers). Sure, the project is barely moving. Sure, it is only marginally better known than the lyrics of Sabbat's "Blood for the Blood God", but it IS kernel-level graphics at its purest.
What, you DON'T use it? You kick up an enormous fuss about what should be done, but haven't actually tried the idea out yourself?
Food no use to White Dwarf. Thrud the Barbarian and some Call of Cthulhu articles would be good, along with a total ban on anything by Games Workshop. That's what caused it to gravitationally collapse in the first place.
It was not unusual, at that time, for artists to use a range of techniques for capturing the key points of the person they were painting, to avoid having a person sit for ages whilst being painted. This means that although the painting would not be a "true" photograph, it could have been extremely close to one.
On a related note, this might also explain the resemblance to Leonardo. Let us say that he did, indeed, have a woman sit for just long enough to sketch in the key facial lines. He would then have needed to add in the skin texture and other features that couldn't have been captured by whatever method he used. It would be logical for him to have used his own face to capture such information. The Mona Lisa would then have been a composite of the original model and himself, which means that it would indeed have a resemblance to him.
X-Ray analysis of the original painting reveals sketches and paintings below the Mona Lisa - though there was no sign of anyone having written "This is a fake" in felt-tip pen, much to the chagrin of Doctor Who fans. It would be interesting to know how the different levels relate to each other - were the earlier pictures earlier versions of the same painting? If they are analyzed with the same software, does it produce the same result?
Vancouver in Washington State or Vancouver, Canada? If Canada, Telus has crap engineers and might well be willing to pay to have people who know what they're doing, but dunno that for sure. If Vancouver WA, then Lightfleet might well be interested - they're on the lookout for high-end developers and I can say for certain that it is going to be tough for them to find the level of skill they want. What's the worst that can happen if you ask?
Binary systems can be near-identical, but don't have to be. That is not a requirement. A star forms when a region of hydrogen/dust cloud has a slightly higher density that normal and the hydrogen collapses inwards under gravity. Alpha Centauri formed from the same hydrogen/dust cloud the sun did but is clearly not identical as it has no viable solar system at this time.
It would be possible (not likely, but possible) for a star to have formed in a position and at a time such that Alpha Centauri prevented it from building to the same mass as the sun. Nothing impossible about that.
Furthermore, several of the known star nursaries that are known are being fed by a massive blast of hydrogen and radiation from a violently dying star. It is probable that some (or all) of these stars will be ripped to shreds, as the forces are simply too great. Again, it is unlikely but not totally impossible the star nursary from which Sol and Alpha Centauri emerged was fed by such a destructive force and that not all of the stars that emerged survived.
There are other oddities - a star was recently seen catapulted from the central core of the galaxy, for example. Stars sometimes steal planets (or even companion stars) from other systems. There are "rogue" supermassive gas giant planets that have left their solar system of origin and are wandering - the sun is big, but it probably wouldn't take kindly to a planet a hundred times the mass of Jupiter doing a belly-flop into it at many times the escape velocity of the solar system.
In short, there's plenty of ways to destroy or otherwise dispose of a star of a paltry one solar mass. Do I think it happened? No. I don't believe Sol ever had a companion star, unless it was Alpha Centauri, but even there, there's a lot of surrounding rock around Alpha Centauri - too much, IMHO, for it to have been a companion to Sol.
Do I think the orbits of the planets are strange? Yes - doubly so because they mostly fit predicted models that are largely based on Keplar's motions modified for Einstein's relativistic laws. Three-body problems are bloody hard, chaotic systems. In our solar system, you really have to solve a four-body or even five-body problem to get useful results for a single planet over any decent timeframe. If you want to model over a five billion year timeframe - necessary to understand how the system reached its current state, you really want to solve a fourteen-body or fifteen-body problem, where not all of those bodies still exist and so we cannot know their mass, initial position or initial velocity.
Short of possibly one of the top 5 supercomputers, I do not believe that there is a computer out there capable of a fifteen-body relativistic simulation with enough granularity to prevent the chaotic variables swamping the results over a five billion year simulation. In consequence, I do not believe that any theory - however "proper" - is meaningful at this point. The quality of the computer models isn't there.
Slashdot Mirror
As Tolkein once noted in his essay "On Fairy Stories", if you have to suspend disbelief, the author failed. The true test of a good story, Tolkein argued, is that you shouldn't have to suspend anything. It should feel real to you, at least to the point where things "just make sense". Lengthy technobabble and explanations shouldn't be necessary.
Again, though, most of this is equally true, no matter what the setting. When the setting is exceptionally difficult, expensive or unusual, therefore, the setting has to be relevent. Asimov's Foundation series could not work in a Wild West setting. The original Star Wars trilogy uses space to express enormity - something no other setting could provide - so whilst the story is a fairly trivial quest in and of itself, it couldn't carry the range of expression in any other setting. The Matrix is also a fairly trivial quest, but requires the deformable, maliable nature of virtual reality in order to cover the metaphysics of the story. It simply couldn't work anywhere else.
Serenity, on the other hand, had nothing new in it and required nothing from the setting. The story looked like a simple mesh of a cowboys-and-indians western with Matrix-style combat, Libertarian vs. The Evil Commies politics, and stole the "alien enemy" from an Asimov short story.
If you want to regard mental processes as a form of sexual reproduction in the most abstract form possible, then I guess it would describe the essential underlying mechanisms involved in how people think, and would certainly go along with a fairly common attitude amongst engineering types that the product of their mind's work is their "baby".
But they trolled the court, they didn't bear it. :)
Actually, even if users didn't check, it would be easy enough for someone like E-Bay to run a promotion targetting ISPs - "if you add our automatic phish frier, you can carry our flashy 'secure e-commerce' logo and be listed as a 'trusted' partner!", where the phish frier simply drops any e-mail that has a FROM claiming to be from E-Bay but where the signature is missing or is incorrect.
ISPs love anything that makes them look good - especially when it doesn't cost them anything. All you'd really need is for a couple of the major players to provide ISPs with such filters, and for just one or two of the major providers (AOL, Comcast, etc) or major webmail providers (Yahoo, Hotmail, Gmail) to install them. I would see that as very doable and would cut down on the phishing scams, even if they didn't eliminate them.
PGP/GPG use RSA or ElGammel public-key encryption to store a secret key, which is then used to decrypt the actual message. This is faster than using RSA or ElGammel for the entire message, but is only as strong as the weakest algorithm in the chain. There are substantially faster public-key algorithms, but they are either known to be broken (HFE), heavily encumbered (HFE, NTRU) or regarded with suspicion (ECC).
The reason this is important is that no authentication scheme can be any better than the method used to prove or validate the identity. (Duh!) Therefore, it is essential that the authentication scheme can be trusted to do what it says - authenticate that the message originated with the person the message claims to be from, with absolutely no possibility of the message originating with anyone else or being modified en-route (except for relay headers).
This means that PGP-style encryption does not prove identity. The message must be signed to prove identity, if the public key is only used to hide a secret key. Even then, with hashing algorithms tumbling like dominos (with or without sparrows helping), you need to be somewhat strict about what method is used for authentication. In four or five years, it is entirely possible to imagine skript kiddies being able to fake MD5-based signatures and for organized crime syndicates to be able to fake SHA1-based signatures. As these are the two largest sources of phishing scams, any approach which they are likely to completely defeat within the lifetime of any standard adopted is useless.
An Internet-based protocol can be considered of having a life-expectancy of 20 years, with no substantial modifications being possible. Maybe 30 years, in some cases, even if the infrastructure is incapable of handling the load. (IPv6 demonstrates that.) That does not mean any given e-mail must be proof against 30 years of concentrated effort. Digital certificates and encryption keys usually expire after a few months or a year, so we don't care if anyone breaks a particular key after that. The ideal signature scheme, then, has only to be reasonably secure for the same few
In Britain, that leaves the House of Lords, who can threaten to veto all legislation from now to eternity. Not a whole lot anyone can do to stop them, either, as they're not elected and it's next to impossible for anyone to impeach them. Still, Tony Blair has threatened to eliminate the entire House (and one can never be quite sure on the interpretation on that) if they defy him too much.
There's also the European Court of Human Rights, but they have only limited powers to enforce their rulings. Generally, though, Governments will bow down to them. At least, in public. The European Court is interesting in all of this, because they are very unpredictable.
The last great hope of Europe is, ironically enough, the British Intelligence network. "Huh?" you might very well say. They have apparently brought down previous British governments (such as that of Harold Wilson), when they have believed the Government had become a danger to Britain. Not a million miles from the way the Watergate scandal broke in the US, only there is evidence that MI5 collectively acted against their superiors in Whitehall, rather than a single individual acting on their own.
Is it even remotely likely MI5 would bring down Tony Blair? Probably not. Would it be even possible? Oh, certainly. And, as noted, there is some basis for believing they've taken matters into their own hands before.
But this is mostly by-the-by. George Bush can't be re-elected and his supporters control all divisions within the Government, so he has no consequences to consider. In Europe, politicians are forever looking over their shoulders. They are always at risk from scandals or other PR disasters. They are therefore far more likely to listen to people, provided it is done right and they consider the risks too great. That is why an organized voice of the IT community must present a coherent case to them. They will listen - they can't afford not to.
It is also important to remember that although data retention laws require information to be available for security reasons, the Data Protection Act prohibits making that information available to anyone else. In fact, most of Europe has incredibly strict privacy laws - along with laws prohibiting the trade of such information to organizations and companies that are outside the jurisdiction of those privacy laws.
A good solid campaign by European technophiles, to remind Euro MPs (and regular MPs) about the British and European privacy laws with regards to personal data, especially when coupled with reminders of Microsoft's extremely dodgy past on security issues, would be likely to derail Microsoft's efforts entirely, as their proposals are technically illegal and politicians in Europe - at least for now - are eager to NOT be seen aiding and abetting lawbreakers. In fact, a solid-enough campaign that also brought in Microsoft's status in Europe as a monopolist guilty of breaking trade laws might potentially finish off Microsoft in Europe entirely.
Before anyone marks me as rabidly anti-Microsoft (I'm not rabid, I just don't like them), this post is not a condemnation of the corporation. It is merely pointing out that their proposal violates EU privacy directives and assorted national laws. This is more likely ignorance than malice, as America has no concept of privacy, but that simply isn't relevent. It would be relatively easy for someone to spin this in a way that would leave politicians with very cold feet. Politicians aren't generally brave - that's not how to get re-elected. Politicians are professional cowards. No sane coward is going to want to be seen breaking the law - or even potentially breaking it - right now. If that were to be how opponents presented it, most politicos would back off very very quickly.
If one of them went back in time and changed things, the spelling might be different.
Well, it is certainly true that a book I have from 1750 describes thunder as the explosions of evaporated gunpowder, so a lack of understanding is certainly possible. On the other hand, we're in the early 21st century now, so I would have expected somebody to have figured this all out in the intervening hundred to hundred and fifty years.
If there was a "simple" solution to this problem, it would have been solved by now. Nonetheless, I do believe that the problem is solvable and that the solution does involve regulation mechanisms. I suspect that the solution must involve some sort of "blind" element (a jury pool system, for example) so that it is not possible for any particular side to know who to bribe. It must also have people with the necessary knowledge but who are unlikely to have been drawn into the system so far as to have become corrupt themselves. The only way you'd get that combination is to introduce a mix of recent graduates and the self-taught. The remainder of the regulatory body should be directly elected as per any other representative, only using a proportional representation system to prevent domination.
It still wouldn't be perfect, but the randomness of one component, the youth of another and the capricious nature of PR elections would make most efforts at corruption extremely difficult and VERY expensive, and therefore probably much reduced.
Now, how does this relate to the lights in the 1800s? Oh, quite easily. I suspect the lights were quite probably fires, but considerably further away and in a completely different place than the observers had expected - which is why they never found anything.
As for people chasing the lights and never reaching them (according to another poster), this is exactly what you expect from an optical illusion from refracted light. Most people have seen this with rainbows, which are also caused by refraction through water droplets. It's the same mechanism, so you get the same "moving" effect. Duh.
In fact, once people had observed they could not "approach" the lights, the physics of it should have been obvious. There aren't many types of illusion which work that way. You can approach a mirage, for example, but it vanishes when you get "too close". If you shine a bright light onto fog, you will get reflected light from it. Etc.
In other words, a shortfall is not a product of a particular society or of outsourcing, it is a simple product of economic forces. Or, at least, those forces that lead to stable economies. Most economies are highly unstable, which means that instead of reaching a dynamic equilibrium, they are highly chaotic. Indeed, much of what is known about chaos theory comes from people studying economics, so there have been fringe benefits.
The instability largely comes from political corruption, monopolies of any kind, insufficient worker protections, price gouging, accounting fraud and other such elements, where the normal feedback mechanisms cannot work or are even actively prevented from working. What you end up with is a positive feedback loop, which leads to the whole boom/bust system that America in particular is infamous for, and which India is likely to lurch into, because it has absolutely no comprehension of the kinds of controls you need to keep the system sane.
The problem with America is that Government regulators are largely financed by the corporations they regulate. This is like having a computer virus scanner asking a virus if it is present. You think it's ever going to tell the truth?
India is in a worse mess, because bribes and back-handers are common practice on the sub-continent. That's like installing a computer virus AS the virus scanner. And sooner or later, jobs will migrate to ever more opaque, ever less accountable, ever more corrupt systems, because those are cheap - in the short-term, at least. Corruption is always expensive in the long-term, just not necessarily for the ones guilty of it.
The only way for India to build a stable economy out of the outsourcing is to have the Government conduct strict, well-defined oversight in a 100% independent fashion. However, that will drive up costs considerably. It will also drive up standards, but accountants don't include standards on the balance sheet and it's the balance sheet that matters. Stability in India will, therefore, effectively end outsourcing there. Which means that India must choose between being popular or being viable. It can't be both.
It means we're opposed to those who are opposed to the opposition of those who oppose the opposition's overlords. I guess that makes it The Story of "O".
Let's see, what other examples are there? Incorrect BIOS writes, as another poster noted, are Bad News, but they are technically recoverable. Well, usually, anyway. If there is nothing to reset the BIOS to, it can be much more troublesome. Hard disk thrashing is a bigger problem, as that can cause an actual head crash.
(So that non-geeks can get some idea of why head crashes are BAD, your typical hard drive has a diameter of 3.8" - and therefore a cicumference of about 12". It revolves 7200 times a minute, which means the outside of the disk is travelling at a speed of 1432.6" per second. Now, imagine an extremely delicate read-head striking that disk. Now picture that scene from Star Trek:TNG where the saucer section from the Enterprise makes an emergency crash landing. That is what your hard drive now looks like.)
Other ways software can cause hardware damage: Loading microcode into the CPU that does the same as CPUburn, only a great deal more savagely. Try googling for Service Pack 6 and you'll find machines that could not be rebooted. At all. They were utterly dead. Buggy drivers on buggy hardware aren't a good combination. If you've an old BBC Micro handy, try the following instruction:
10 DIM a%(1,1,1,1,1,1,1,1,1,1,1,1,1,1)
The Commodore PET allowed you to alter video timings, but was never designed for such modifications. It was therefore possible to cause permanent damage to the display.
These are only a few examples, but computer history is littered with them. The Manchester Mk 1 had far more opcodes than instructions, with unused opcodes running through utterly unpredictable paths in hardware. I think it used 32-bit addresses, but it only had 32 words of memory. Again, going outside of physical memory was unpredictable. You programmed that machine with EXTREME care.
However, it is highly inefficient to modify data in a purely serial manner. Serial went out with punch tape. You are always going to have to slice data in some form or other, because data will typically have far more dimensions to it than can be entered or displayed.
The problem is, most slicing mechanisms work on instantaneous snapshots. They are "horizontal" slices, if you like, across a moment in time. But flow-based programs don't just have an instantaneous state and therefore don't have any reason to be limited to just receiving data as if that was all the program could handle. Flow-based programs are built around the system/model as a whole, over all possible paths and eventualities. Feeding data into them a single time-slice or event-slice at a time is no more efficient than writing a program to plot a sine wave by manually entering fixed values at fixed points.
Let us take this one step further. Not only have people broken away from the instantaneous state, they have broken away from sequential ordering. Programmers, for the past twenty to twenty-five years have been much more interested in multi-threaded, multi-data, anticipatory, event-driven programs and computer architectures.
The only "sane" way to handle the I/O for such software is by allowing time to be an element in the display. And this is fine. Humans are great at recognizing patterns - it is what our brains are optimized for. And there are no patterns in a single slice, because by taking a single slice, there is nothing to relate to and therefore no patterns to observe.
The only interest manufacturers and developers have had in 3D displays, though, is allowing people to display a more "complete" or sophisticated slice. They're still completely missing the point that interactive computing isn't batch processing on tranquilizers. If you're going to have interactive computing, make use of having a human at the console. Give the human patterns! Allow them to manipulate those patterns directly. In other words, allow them to directly manipulate the change of the data, not just the value of the data.
If you're going to have "complete" slices via a 3D display, then ergo, you must need a 4D display to manipulate the way the data changes, particularly if you want to be able to see how the flows split and recombine, as a single slice cannot capture information about the path(s) you are NOT on. Furthermore, to go back to the sine example, slices are only good for discrete events, they're not useful for continuous systems. However, 99.9% of reality is continuous. (Which is why computer models are generally crap. Discrete approximations of continuous systems are almost never going to be any good.)
High-performance computing has reached the limits of sliced-and-diced discrete-event I/O formats. If HPC is to progress much further in human-computer interfaces, they will have to display change as something more than just frames in an animation. It isn't realistic to the user, isn't logical to the program and isn't producing useful results anyway. Since it is generally the HPC labs that have the Really Big Money, this is where you will see any new display technology.
LCD monitors that can use polarized light to display full-color 3D images - hey, that's cool. The systems at SC|05 were most impressive for what they were. But they're not system images, they're still snapshots. People want to manipulate systems and to do that they will need to "see" systems and interact directly with them. And 3D just isn't enough. You need 4D for that, whatever 4D might actually mean in a p
If you look at Microsoft's scorecard for service packs, sure, you can find even-numbered ones that locked systems solid. You can find odd-numbered ones that did that, too. The only method I can suggest for safely installing a service pack is to wait until someone publishes a regression test on that SP on the machine you use with the hardware you have. The next-best is to find a combination of tests which cover the arrangement, but that's not as good, as interactions between components can be critical.
The third option is to make extensive backups, then try installing the service pack. If it doesn't work, restore from backup. The problem with this is that there are rare bugs that cause actual hardware damage. I only recall reports of such damage for one Windows service pack, but I am definitely unhappy about the fact that there are any such reports at all. It is for that reason I relegate this to being a third option, because (IMHO) it is preferable to let someone else fry their computer - doubly so as Microsoft won't take responsibility and manufacturer warranties are usually voided if you change the system at all (such as installing service packs).
On the other hand, I know that many early machine developers at Manchester University, where the first stored-program computer was built, remained on the faculty until they retired. I also know people from Imperial Computers Limited (a UK mainframe company) who were extremely well-off.
The serious answer, then, is that it varies.
As for some of the not-so-serious answers others have offered... well, I always was suspicious of University caffeteria food...
The cards were rectangular, the chads were roundish and the legal case went in circles. Nothing square about it.
What, you DON'T use it? You kick up an enormous fuss about what should be done, but haven't actually tried the idea out yourself?
Food no use to White Dwarf. Thrud the Barbarian and some Call of Cthulhu articles would be good, along with a total ban on anything by Games Workshop. That's what caused it to gravitationally collapse in the first place.
They went to a comic/magazine collector and asked them how much that issue weighed.
On a related note, this might also explain the resemblance to Leonardo. Let us say that he did, indeed, have a woman sit for just long enough to sketch in the key facial lines. He would then have needed to add in the skin texture and other features that couldn't have been captured by whatever method he used. It would be logical for him to have used his own face to capture such information. The Mona Lisa would then have been a composite of the original model and himself, which means that it would indeed have a resemblance to him.
X-Ray analysis of the original painting reveals sketches and paintings below the Mona Lisa - though there was no sign of anyone having written "This is a fake" in felt-tip pen, much to the chagrin of Doctor Who fans. It would be interesting to know how the different levels relate to each other - were the earlier pictures earlier versions of the same painting? If they are analyzed with the same software, does it produce the same result?
Vancouver in Washington State or Vancouver, Canada? If Canada, Telus has crap engineers and might well be willing to pay to have people who know what they're doing, but dunno that for sure. If Vancouver WA, then Lightfleet might well be interested - they're on the lookout for high-end developers and I can say for certain that it is going to be tough for them to find the level of skill they want. What's the worst that can happen if you ask?
It would be possible (not likely, but possible) for a star to have formed in a position and at a time such that Alpha Centauri prevented it from building to the same mass as the sun. Nothing impossible about that.
Furthermore, several of the known star nursaries that are known are being fed by a massive blast of hydrogen and radiation from a violently dying star. It is probable that some (or all) of these stars will be ripped to shreds, as the forces are simply too great. Again, it is unlikely but not totally impossible the star nursary from which Sol and Alpha Centauri emerged was fed by such a destructive force and that not all of the stars that emerged survived.
There are other oddities - a star was recently seen catapulted from the central core of the galaxy, for example. Stars sometimes steal planets (or even companion stars) from other systems. There are "rogue" supermassive gas giant planets that have left their solar system of origin and are wandering - the sun is big, but it probably wouldn't take kindly to a planet a hundred times the mass of Jupiter doing a belly-flop into it at many times the escape velocity of the solar system.
In short, there's plenty of ways to destroy or otherwise dispose of a star of a paltry one solar mass. Do I think it happened? No. I don't believe Sol ever had a companion star, unless it was Alpha Centauri, but even there, there's a lot of surrounding rock around Alpha Centauri - too much, IMHO, for it to have been a companion to Sol.
Do I think the orbits of the planets are strange? Yes - doubly so because they mostly fit predicted models that are largely based on Keplar's motions modified for Einstein's relativistic laws. Three-body problems are bloody hard, chaotic systems. In our solar system, you really have to solve a four-body or even five-body problem to get useful results for a single planet over any decent timeframe. If you want to model over a five billion year timeframe - necessary to understand how the system reached its current state, you really want to solve a fourteen-body or fifteen-body problem, where not all of those bodies still exist and so we cannot know their mass, initial position or initial velocity.
Short of possibly one of the top 5 supercomputers, I do not believe that there is a computer out there capable of a fifteen-body relativistic simulation with enough granularity to prevent the chaotic variables swamping the results over a five billion year simulation. In consequence, I do not believe that any theory - however "proper" - is meaningful at this point. The quality of the computer models isn't there.
They were going to send a Clipper ship, but someone got confused and packaged the astronauts up in a Clipper chip instead.