Of course not. That's a $26,000 triple-autoclaved, platinum-coated, Swiss optic mirror, affixed to a $43,424 surgical steel stick, made to NASA's exacting specifications down to the micrometer. The 3M company launched a whole new division to create the special cellphane tape (release strength 3.434 KPa +/-.002 KPa), $113,285 per yard (but they only used about six inches; the rest is being used to tape the fragments together).
I wonder if they plan to blacklist certain payees. "Hey, do you really want to send $500 to GamblingOnline.com and therefore risk a heavy fine if we find out that they're a gambling site? Click 'yeah, I'll take a gamble' to continue." At the very least, inform you that other users have been fined for sending money to that payee.
It's certainly their right to enforce whatever contract terms you agree to, but I'd consider it unpleasantly sneaky if they didn't warn you very explicitly first, at least when you're going to a known porn or gambling site. That is clearly, at the time you make the payment, not just in the fine print along with a hundred other regulations when you sign up. That's not a legal requirement, but if they're planning to use this as a source of significant income rather than an ass-covering gesture it would tip them into the Evil-Never-Use category for me.
Thanks for the interesting analysis. The article sez, "An analysis of its emissions found it contains water, which suggests its mass is in the range of planets rather than stars." (That's from the article; for some reason the preprint didn't render for me.)
If the red object is a star, could it contain water? Or something in the spectral lines that might look like water?
It certainly doesn't look like what I'd expect a planet to look like in a photo (not that I'm an astronomer or anything). It's awfully bright to be a reflective object. Maybe that photo is retouched?
Those are all good karma whoring advice points, but I don't think they do too much damage to the system. It keeps away the worst of the trolls.
I worry about a WhoreBot which uses Eliza-type mechanisms, and pulls text from highly-moderated comments in similar discussions (perhaps from dups) to create high-karma accounts with no human intervention. The owner could then do ugly things with mod points, post offensive things with the +1 modifier, and generally make a real nuisance of himself.
Not to mention moderate offtopic posts like this mine into oblivion.
It's easier if you own your own domain. But even with your gmail address, you can compose new addresses on the fly like foo+company@gmail.com. (That's a mail-standard thing, not a gmail-specific thing.)
Gmail doesn't allow you to configure filters; I'd love to tell it "reject all emails from me+UnscrupulousCorp@gmail.com". But you can do it with a domain over which you have more control.
Unfortunately, this won't work indefinitely: the spammers will eventually just remove everything after the plus sign. There are options after that, but it gets less convenient.
The trick is, what do you replace it with? There are a lot of design constraints on email, among them:
* Sending message should be free or extremely cheap * It should not be required to receive an invitation to talk to somebody
You can quibble with those requirements if you want to design a new system, but if you follow them any system you propose risks being spam-ridden. The spammers will not say, "Oh, gee, they've all moved to a different port and protocol, let's forget it then." They'll adopt any new protocol, faster than users will.
So what about present email are you willing to give up? Converting from "free" to "extremely cheap" sounds promising, but it's still prone to the army of zombies, and exchanging trivial amounts of cash is still difficult and expensive.
There are various ways to introduce blocks in the "anybody can talk to anybody" system. Some systems email you back when you send me a message for the first time, which at least proves the existence of a back path and to a small degree a real human (not a zombie) on the other end. Bayesian filters provide extra points to people who have emailed you before without excluding people you've never heard of.
Or maybe we weaken the second requirement by distinguishing between promiscuous and non-promiscuous addresses. My friends email me at one account, and if I could I'd give each of them a separate address. People I trust less get different accounts. People who break the trust find that the address disappears, and because those addresses aren't promiscuous, relatively few other people are inconvenienced by that. I've effectively whitelisted those addresses.
But I also monitor info@foo.com email addresses, which really do want to take email from anybody in the world. I can't drop those when they get spammed, because many people are expecting to get to me through them. But if we made promiscuous addresses rare, we could use more whitelists and perhaps change the balance.
Perhaps if your average spam-buying-jackass@comcast.net were able to receive mail only from people he'd whitelisted, he'd get less spam and the spammers would give up. But that would be wildly inconvenient for him.
The point is, most of these could be built on top of SMTP, and any SMTP alternative you propose is going to have either promiscuity or conveninence problems. Just dropping SMTP just moves the problem to a new protocol but with massive infrastructure pain.
If they are authentic, then it's just a bizarre coincidence that they look so much like Word documents. I'm not saying that's wrong; I'm simply marvelling.
As you say, there's no reason for the White House to concur with the documents if they were forged, and there's no reason for them to forge them. So they are probably authentic, on those grounds alone, aside from any work CBS did to get expert authentication. It's just one really weird coincidence that the letter spacings should line up so well, and that a relatively uncommon operation like superscript-th should have been used. (Most people today wouldn't do it if MS Word didn't do it for them.)
That would make sense. I've heard that they go even further and give poor service to people who are really unprofitable to encourage them to quit, but that's just anecdotal. I've been very happy and I rent about 7 discs a month (which means I about break even with respect to going to Blockbuster, minus the late fees.)
I've done the same thing, and they look about the same to me. I downconverted it to 72 DPI, and they're still not identical, but I'd believe that they were if the document were faxed.
It seems an extreme coincidence that they should be so similar, and yet I can't figure out why it should be done. The document is detrimental to Bush, but the White House was the one who released the document.
Weird, weird. Best guess is that somebody in the White House is covering up something even worse, but that doesn't make a whole lot of sense either.
People file FOIA requests all the time, successfully. Just the most recent example off the top of my head was more Bush guard unit records, and there are many many more.
I believe the present government is too secretive, and I can't swear the Secret Service has never shown up at somebody's door to politely suggest cancelling a request, but the act is used all the time.
Ya know, I was an immense Star Trek fan back when there were 79 episodes, and I had every single one of them memorized. I won trivia contests. I watched it as much as three hours a day.
Somehow, the movies got in the way of my perfect knowledge of the series. And then TNG added scores of new episodes, more than the original series.
And, maybe, I stopped being an adolescent loser (and became a post-adolescent loser). But I still want to blame my falling out of love with Star Trek on there being too much Star Trek, not all of it lovable.
Actually, I'm not convinced that Nader will draw more Democratic voters than Republican ones.
In 2000 he represented a more liberal choice that many Democrats found attractive, especially with their particularly uninteresting candidate. Nader convinced people that Gore and Bush presented no real differences, and the candidates didn't do much to fight that perception: they were both boring.
Four years later the Democratic candidate is not significantly more interesting, but the opponent is far more unattractive to Democratic voters. For Democrats, its not simply the unfortunate opportunity for Republicans to control the government any more; it's now a matter literally of life and death.
Therefore, Democrats line up behind Kerry. But many Republicans are uneasy about Bush, both on privacy and budget issues as well as the more obvious Iraq question. But there's no way they're going to vote for Kerry; Republicans are far more likely to believe the accusations of the SBVfT (I just can't bring myself to spell out their name). Some will vote for Nader as a no-vote vote.
So at this point, I believe that Nader may well be swinging the race Democratically to the degree that he makes a difference at all. The most recent Gallup poll supports this notion: the three-way race is 52/41/3 (Bush/Kerry/Nader), and the two-way is 54/43 (Bush/Kerry). In other words, Nader draws equally from both sides.
That's just general vote, of course. In 2000 Nader's dropping out needed to have swung less than 600 votes to change the outcome, due to the electoral college. But with the numbers I see, that ax may swings both ways in 2004.
Uh, yeah. China is a NWS (Nuclear Weapons State) under the Nuclear Non-Proliferation Treaty. So they get to have nukes, along with France, Russia, the US and the UK.
Everybody else has to forgo, except that India, Pakistan, and Israel never signed the treaty, and North Korea, who did but has nukes anyway.
True. But factoring is easier to explain in a short example than TSP or (God forbid) 3-SAT.
Re:Nope, wrong, invalid.. nothing to see here.
on
The End of Encryption?
·
· Score: 4, Informative
Yes, but there's a reason P=NP is of particular interest for crypto problems.
NP problems as a category are easy to check answers, but hard to compute those answers. So a whole category of challenge-response systems are possible. I use the easy (checking) side of the NP problem and make codebreakers use the hard (computing) side.
For example, it's hard to factor large composite numbers, but easy to check if a set of primes multiplies out to that number.
Sure, there are plenty of other categories of crypto, but they're harder to deal with. One-time pads are hard to distribute, and quantum mechanical stuff isn't ready yet. But public-private key cryptosystems are based on computations like factoring: it's easy to encrypt something based on the large composite number, but harder to decrypt it unless you have the factors at hand. So I can distribute the large composite number (so anybody can send secret documents to me), and be fairly sure nobody will break the crypto.
Unless P=NP, in which case factoring the number will also be easy, and we'll have to resort to something smarter, like quantum crypto (assuming it can be made to work practically).
Yep. But it means that every Blu-Ray DVD player you buy gives Microsoft its inch of green. You won't be allowed to buy a Blu-Ray DVD player without it, even if you never play any VC-1 coded movies.
I dunno what the licensing terms for VC-1 are. I suspect they're not too burdensome, a few cents on every DVD player (like the mandatory Macrovision code you're currently buying with every DVD player). I'd like to think the consortium worked it out before confirming the standard, rather than giving Bill Gates the opportunity to appear on every TV screen in the world and say, "Ha ha! You pitiful fools! The licensing for VC-1 will be one skrillion dollars."
From a business standpoint, no large company is ever satisfied with where it is. The theory is that you have to grow or die.
If you were a Microsoft stockholder, would you want the company to look at the pile of money out there in the music business and say, "No thanks, I couldn't possibly eat another bite, I'm full"?
Companies exist to make money for their shareholders. That's the only reason they exist. Unlike a person, the company doesn't have a conscience. There are things you wouldn't do for money, but a company has no pride or dignity, except to the degree that compromising them costs it money.
That is, Microsoft may be able to do great business shooting puppies, but it would cost them a lot more money since people would avoid their other non-puppy-shooting businesses. That's the only reason they wouldn't do it. In a less stupid example, Microsoft could probably buy Apple outright, but it's constrained by a government. They're never, ever constrained by conscience, only by the government and the market.
So in the upshot, no company is ever satisfied. If it were, its shareholders would get upset and replace its board with one who wasn't satisfied. A few companies, most notably Ben and Jerry's, manage to find other models for business than simply money, but they're few and far between. Money is a hell of a motivator.
True, but the point is clear: the longest carbon nanotube structures are still seven or eight orders of magnitude too short, and the longest ones with sufficient strength for a space elevator are another order of magnitude on top of that.
That's such a massive leap that I'd like to see people aim at more practical structures, which are still far from a given, before turning their heads to something that may prove too difficult even if the materials science can keep up with it.
I just saw an ad at Yahoo for HP + iPod. They figure there's cross-branding to be done. Apple's taking advantage of a marketing channel they never had before. HP gets to sell a technology to many who have heard of the iPod but were buying Dells instead.
This is not a technical advance, it's a marketing advance. Marketing is a rather arcane science, and many of those who practice it are fools. I'm not a marketer, so I can't tell you if it's a good marketing idea or a bad one. But the answer to any question beginning with "Why" is always "money".
WinFS is an interesting, bold, and novel take on a file system, but I'm not sure why it's taking so long for them to implement. They've been working on it for a very long time. It's complicated, but it doesn't seem ten-years-by-a-dedicated-team complicated. I can't help but think that once Microsoft comes out with a reference model, there will be an open source reimplementation in months.
Microsoft has higher demands on it, and it's harder to develop it the first time, and presumably their implementation is optimized to within in an inch of its life, but I still don't see why a project they're working on now won't be ready for 2006.
Could it be that they want to adapt their applications to use the new features before they release it? That I could see taking forever, since everything from Word down to the format Spider Solitaire saves its games in would be affected. But I assume that they've implemented a Win32 filesystem API on top of it, and presumably with tolerable performance, so why not release it and adapt the apps later?
The Ansari X-prize seeks to reproduce an effort that had already succeeded, and been substantially surpassed, by several governments.
A "space elevator", on the other hand, is totally unlike anything ever done before. As I read in a Slashdot post some years ago (referring to nanotubes, the favorite among space-elevator aficionados), "When somebody has built a 40,000 millimeter bridge across a creek on campus, then we can start to talk about a 40,000 kilometer bridge straight up".
The fact that we have not yet achieved one millionth of the task (and in fact fall several orders of magnitude for that) suggests to me that, much as I would love to see a space elevator in place, the job today belongs to materials scientists who are looking at shorter-term goals.
An eye to the future is great, but experimenting on climbers is like practicing the high jump: if you're jumping twice as high today as last year, I wouldn't start drawing any exponential curves. The ribbon is the really, really hard part, and we're currently so far away from it that research energy is better spent elsewhere for a while. 2010 is way, way too close.
Maybe with enough motivation we could get that 40,000 mm bridge by 2010, but somehow I doubt you're going to raise $10 million to build a bridge. The X-prize shot somebody into space for that kind of money.
I'm prepared to be wrong. I'm a software developer, and I've learned that as a consultant I can say, "Your project is doomed" with 95% accuracy before I've even heard your name. Being a nay-sayer is easy. But the real trick is being able to spot the 5% that will actually be profitable, and there are a lot of projects more immediately deserving of this kind of money.
There are a million reasons why it makes things complicated. But that doesn't mean it was a bad idea. Especially at the time, on those 33 MHz computers with 640K of RAM. Premature optimization is the root of all evil, but you really can't wait for Moore's law to catch up with you before you ship the product with acceptable performance.
Ah, the 1999s. They should be peaking in flavor just about now. I'm so glad we didn't have to eat all of them back in 2000; they were so immature then. With an extra four years in the package to ripen and mature, the aromas will be brighter and the tannins will have mellowed.
Yes, but only because the prediction was made in the first place and scared off the terrorists. We owe Mr. Gostev a huge debt of gratitude no matter what happens.
Slashdot Mirror
Of course not. That's a $26,000 triple-autoclaved, platinum-coated, Swiss optic mirror, affixed to a $43,424 surgical steel stick, made to NASA's exacting specifications down to the micrometer. The 3M company launched a whole new division to create the special cellphane tape (release strength 3.434 KPa +/- .002 KPa), $113,285 per yard (but they only used about six inches; the rest is being used to tape the fragments together).
I wonder if they plan to blacklist certain payees. "Hey, do you really want to send $500 to GamblingOnline.com and therefore risk a heavy fine if we find out that they're a gambling site? Click 'yeah, I'll take a gamble' to continue." At the very least, inform you that other users have been fined for sending money to that payee.
It's certainly their right to enforce whatever contract terms you agree to, but I'd consider it unpleasantly sneaky if they didn't warn you very explicitly first, at least when you're going to a known porn or gambling site. That is clearly, at the time you make the payment, not just in the fine print along with a hundred other regulations when you sign up. That's not a legal requirement, but if they're planning to use this as a source of significant income rather than an ass-covering gesture it would tip them into the Evil-Never-Use category for me.
Thanks for the interesting analysis. The article sez, "An analysis of its emissions found it contains water, which suggests its mass is in the range of planets rather than stars." (That's from the article; for some reason the preprint didn't render for me.)
If the red object is a star, could it contain water? Or something in the spectral lines that might look like water?
It certainly doesn't look like what I'd expect a planet to look like in a photo (not that I'm an astronomer or anything). It's awfully bright to be a reflective object. Maybe that photo is retouched?
Those are all good karma whoring advice points, but I don't think they do too much damage to the system. It keeps away the worst of the trolls.
I worry about a WhoreBot which uses Eliza-type mechanisms, and pulls text from highly-moderated comments in similar discussions (perhaps from dups) to create high-karma accounts with no human intervention. The owner could then do ugly things with mod points, post offensive things with the +1 modifier, and generally make a real nuisance of himself.
Not to mention moderate offtopic posts like this mine into oblivion.
It's easier if you own your own domain. But even with your gmail address, you can compose new addresses on the fly like foo+company@gmail.com. (That's a mail-standard thing, not a gmail-specific thing.)
Gmail doesn't allow you to configure filters; I'd love to tell it "reject all emails from me+UnscrupulousCorp@gmail.com". But you can do it with a domain over which you have more control.
Unfortunately, this won't work indefinitely: the spammers will eventually just remove everything after the plus sign. There are options after that, but it gets less convenient.
The trick is, what do you replace it with? There are a lot of design constraints on email, among them:
* Sending message should be free or extremely cheap
* It should not be required to receive an invitation to talk to somebody
You can quibble with those requirements if you want to design a new system, but if you follow them any system you propose risks being spam-ridden. The spammers will not say, "Oh, gee, they've all moved to a different port and protocol, let's forget it then." They'll adopt any new protocol, faster than users will.
So what about present email are you willing to give up? Converting from "free" to "extremely cheap" sounds promising, but it's still prone to the army of zombies, and exchanging trivial amounts of cash is still difficult and expensive.
There are various ways to introduce blocks in the "anybody can talk to anybody" system. Some systems email you back when you send me a message for the first time, which at least proves the existence of a back path and to a small degree a real human (not a zombie) on the other end. Bayesian filters provide extra points to people who have emailed you before without excluding people you've never heard of.
Or maybe we weaken the second requirement by distinguishing between promiscuous and non-promiscuous addresses. My friends email me at one account, and if I could I'd give each of them a separate address. People I trust less get different accounts. People who break the trust find that the address disappears, and because those addresses aren't promiscuous, relatively few other people are inconvenienced by that. I've effectively whitelisted those addresses.
But I also monitor info@foo.com email addresses, which really do want to take email from anybody in the world. I can't drop those when they get spammed, because many people are expecting to get to me through them. But if we made promiscuous addresses rare, we could use more whitelists and perhaps change the balance.
Perhaps if your average spam-buying-jackass@comcast.net were able to receive mail only from people he'd whitelisted, he'd get less spam and the spammers would give up. But that would be wildly inconvenient for him.
The point is, most of these could be built on top of SMTP, and any SMTP alternative you propose is going to have either promiscuity or conveninence problems. Just dropping SMTP just moves the problem to a new protocol but with massive infrastructure pain.
If they are authentic, then it's just a bizarre coincidence that they look so much like Word documents. I'm not saying that's wrong; I'm simply marvelling.
As you say, there's no reason for the White House to concur with the documents if they were forged, and there's no reason for them to forge them. So they are probably authentic, on those grounds alone, aside from any work CBS did to get expert authentication. It's just one really weird coincidence that the letter spacings should line up so well, and that a relatively uncommon operation like superscript-th should have been used. (Most people today wouldn't do it if MS Word didn't do it for them.)
That would make sense. I've heard that they go even further and give poor service to people who are really unprofitable to encourage them to quit, but that's just anecdotal. I've been very happy and I rent about 7 discs a month (which means I about break even with respect to going to Blockbuster, minus the late fees.)
I've done the same thing, and they look about the same to me. I downconverted it to 72 DPI, and they're still not identical, but I'd believe that they were if the document were faxed.
It seems an extreme coincidence that they should be so similar, and yet I can't figure out why it should be done. The document is detrimental to Bush, but the White House was the one who released the document.
Weird, weird. Best guess is that somebody in the White House is covering up something even worse, but that doesn't make a whole lot of sense either.
People file FOIA requests all the time, successfully. Just the most recent example off the top of my head was more Bush guard unit records, and there are many many more.
I believe the present government is too secretive, and I can't swear the Secret Service has never shown up at somebody's door to politely suggest cancelling a request, but the act is used all the time.
Ya know, I was an immense Star Trek fan back when there were 79 episodes, and I had every single one of them memorized. I won trivia contests. I watched it as much as three hours a day.
Somehow, the movies got in the way of my perfect knowledge of the series. And then TNG added scores of new episodes, more than the original series.
And, maybe, I stopped being an adolescent loser (and became a post-adolescent loser). But I still want to blame my falling out of love with Star Trek on there being too much Star Trek, not all of it lovable.
Actually, I'm not convinced that Nader will draw more Democratic voters than Republican ones.
In 2000 he represented a more liberal choice that many Democrats found attractive, especially with their particularly uninteresting candidate. Nader convinced people that Gore and Bush presented no real differences, and the candidates didn't do much to fight that perception: they were both boring.
Four years later the Democratic candidate is not significantly more interesting, but the opponent is far more unattractive to Democratic voters. For Democrats, its not simply the unfortunate opportunity for Republicans to control the government any more; it's now a matter literally of life and death.
Therefore, Democrats line up behind Kerry. But many Republicans are uneasy about Bush, both on privacy and budget issues as well as the more obvious Iraq question. But there's no way they're going to vote for Kerry; Republicans are far more likely to believe the accusations of the SBVfT (I just can't bring myself to spell out their name). Some will vote for Nader as a no-vote vote.
So at this point, I believe that Nader may well be swinging the race Democratically to the degree that he makes a difference at all. The most recent Gallup poll supports this notion: the three-way race is 52/41/3 (Bush/Kerry/Nader), and the two-way is 54/43 (Bush/Kerry). In other words, Nader draws equally from both sides.
That's just general vote, of course. In 2000 Nader's dropping out needed to have swung less than 600 votes to change the outcome, due to the electoral college. But with the numbers I see, that ax may swings both ways in 2004.
And South Korea too, apparently.
Uh, yeah. China is a NWS (Nuclear Weapons State) under the Nuclear Non-Proliferation Treaty. So they get to have nukes, along with France, Russia, the US and the UK.
Everybody else has to forgo, except that India, Pakistan, and Israel never signed the treaty, and North Korea, who did but has nukes anyway.
True. But factoring is easier to explain in a short example than TSP or (God forbid) 3-SAT.
Yes, but there's a reason P=NP is of particular interest for crypto problems.
NP problems as a category are easy to check answers, but hard to compute those answers. So a whole category of challenge-response systems are possible. I use the easy (checking) side of the NP problem and make codebreakers use the hard (computing) side.
For example, it's hard to factor large composite numbers, but easy to check if a set of primes multiplies out to that number.
Sure, there are plenty of other categories of crypto, but they're harder to deal with. One-time pads are hard to distribute, and quantum mechanical stuff isn't ready yet. But public-private key cryptosystems are based on computations like factoring: it's easy to encrypt something based on the large composite number, but harder to decrypt it unless you have the factors at hand. So I can distribute the large composite number (so anybody can send secret documents to me), and be fairly sure nobody will break the crypto.
Unless P=NP, in which case factoring the number will also be easy, and we'll have to resort to something smarter, like quantum crypto (assuming it can be made to work practically).
Yep. But it means that every Blu-Ray DVD player you buy gives Microsoft its inch of green. You won't be allowed to buy a Blu-Ray DVD player without it, even if you never play any VC-1 coded movies.
I dunno what the licensing terms for VC-1 are. I suspect they're not too burdensome, a few cents on every DVD player (like the mandatory Macrovision code you're currently buying with every DVD player). I'd like to think the consortium worked it out before confirming the standard, rather than giving Bill Gates the opportunity to appear on every TV screen in the world and say, "Ha ha! You pitiful fools! The licensing for VC-1 will be one skrillion dollars."
From a business standpoint, no large company is ever satisfied with where it is. The theory is that you have to grow or die.
If you were a Microsoft stockholder, would you want the company to look at the pile of money out there in the music business and say, "No thanks, I couldn't possibly eat another bite, I'm full"?
Companies exist to make money for their shareholders. That's the only reason they exist. Unlike a person, the company doesn't have a conscience. There are things you wouldn't do for money, but a company has no pride or dignity, except to the degree that compromising them costs it money.
That is, Microsoft may be able to do great business shooting puppies, but it would cost them a lot more money since people would avoid their other non-puppy-shooting businesses. That's the only reason they wouldn't do it. In a less stupid example, Microsoft could probably buy Apple outright, but it's constrained by a government. They're never, ever constrained by conscience, only by the government and the market.
So in the upshot, no company is ever satisfied. If it were, its shareholders would get upset and replace its board with one who wasn't satisfied. A few companies, most notably Ben and Jerry's, manage to find other models for business than simply money, but they're few and far between. Money is a hell of a motivator.
True, but the point is clear: the longest carbon nanotube structures are still seven or eight orders of magnitude too short, and the longest ones with sufficient strength for a space elevator are another order of magnitude on top of that.
That's such a massive leap that I'd like to see people aim at more practical structures, which are still far from a given, before turning their heads to something that may prove too difficult even if the materials science can keep up with it.
I just saw an ad at Yahoo for HP + iPod. They figure there's cross-branding to be done. Apple's taking advantage of a marketing channel they never had before. HP gets to sell a technology to many who have heard of the iPod but were buying Dells instead.
This is not a technical advance, it's a marketing advance. Marketing is a rather arcane science, and many of those who practice it are fools. I'm not a marketer, so I can't tell you if it's a good marketing idea or a bad one. But the answer to any question beginning with "Why" is always "money".
WinFS is an interesting, bold, and novel take on a file system, but I'm not sure why it's taking so long for them to implement. They've been working on it for a very long time. It's complicated, but it doesn't seem ten-years-by-a-dedicated-team complicated. I can't help but think that once Microsoft comes out with a reference model, there will be an open source reimplementation in months.
Microsoft has higher demands on it, and it's harder to develop it the first time, and presumably their implementation is optimized to within in an inch of its life, but I still don't see why a project they're working on now won't be ready for 2006.
Could it be that they want to adapt their applications to use the new features before they release it? That I could see taking forever, since everything from Word down to the format Spider Solitaire saves its games in would be affected. But I assume that they've implemented a Win32 filesystem API on top of it, and presumably with tolerable performance, so why not release it and adapt the apps later?
The Ansari X-prize seeks to reproduce an effort that had already succeeded, and been substantially surpassed, by several governments.
A "space elevator", on the other hand, is totally unlike anything ever done before. As I read in a Slashdot post some years ago (referring to nanotubes, the favorite among space-elevator aficionados), "When somebody has built a 40,000 millimeter bridge across a creek on campus, then we can start to talk about a 40,000 kilometer bridge straight up".
The fact that we have not yet achieved one millionth of the task (and in fact fall several orders of magnitude for that) suggests to me that, much as I would love to see a space elevator in place, the job today belongs to materials scientists who are looking at shorter-term goals.
An eye to the future is great, but experimenting on climbers is like practicing the high jump: if you're jumping twice as high today as last year, I wouldn't start drawing any exponential curves. The ribbon is the really, really hard part, and we're currently so far away from it that research energy is better spent elsewhere for a while. 2010 is way, way too close.
Maybe with enough motivation we could get that 40,000 mm bridge by 2010, but somehow I doubt you're going to raise $10 million to build a bridge. The X-prize shot somebody into space for that kind of money.
I'm prepared to be wrong. I'm a software developer, and I've learned that as a consultant I can say, "Your project is doomed" with 95% accuracy before I've even heard your name. Being a nay-sayer is easy. But the real trick is being able to spot the 5% that will actually be profitable, and there are a lot of projects more immediately deserving of this kind of money.
There are a million reasons why it makes things complicated. But that doesn't mean it was a bad idea. Especially at the time, on those 33 MHz computers with 640K of RAM. Premature optimization is the root of all evil, but you really can't wait for Moore's law to catch up with you before you ship the product with acceptable performance.
Ah, the 1999s. They should be peaking in flavor just about now. I'm so glad we didn't have to eat all of them back in 2000; they were so immature then. With an extra four years in the package to ripen and mature, the aromas will be brighter and the tannins will have mellowed.
Yes, but only because the prediction was made in the first place and scared off the terrorists. We owe Mr. Gostev a huge debt of gratitude no matter what happens.