Um, I really don't think we want lots of people able to develop biological weapons in their basement. We already have enough problems with script kiddies making computer viruses, you'd think they'd learn.
This may be one of those technologies which creates a problem, the resolution of which is that the civilization making it gets knocked back to where it can no longer make the technology. (Classic examples from Science Fiction include certain general-purpose teleporters, as discussed in Niven's classic "On the Theory and Practice of Teleportation", and to a lesser degree the time viewer in Asimov's "The Dead Past".) I suppose that's one solution to the Fermi Paradox....
Radio Frequency interference is a bad thing in airplanes; while not ruling out trying any of those RF options, I'd be a little cautious in risking wireless in a space station if I'm going to be breathing on it.
True, the slashdot article link is to the torture droid and not the Training Remote, but in Dubya's brave new world, IT-0 may be what NASA had in mind after all-- gotta keep close watch on them astronauts. Shifty, the lot of 'em.
[W]hy not have Sun, IBM, Red Hat, SuSe and whoever else is pissed at SCO get together and spend 78 million and buy the bastards?
It's a matter of principle: "If you will not set a good example, you shall serve as a terrible warning." Or, more specifically here, IBM's version of it: "If you sue IBM, we will destroy you." Doing so serves to discourage "sue Big Blue" as an exit strategy for other failing companies-- a corollary of the old principle about Danegeld. Besides, IBM has most of these lawyers on salary... it's good to keep them in practice, just like it's good to keep your knives sharpened. =)
Unsubstatiated rumor has it that members of IBM's legal team have been informed that they will be considered to have failed if SCO even exists after this lawsuit ends.
The article parent links to discusses a separate motion to delay depositions--which, yes, was denied-- not delay the overall trial schedule. The motion to bifurcate the trial, and the motion to extend the Discovery/Deposition period before trial, are reported as being taken under advisement.
Yes, yes, quantum encryption precludes interception; ergo, unlike with IPSEC, "Eve" can't duplicate the QE message during its transfer, store the encoded message for 50 years, and then crack the code with Any Sufficiently Advanced Technological Improvement. So yes, it's useless for protecting storage-- as I noted, the plaintext on either end is still vulnerable-- but it does provide an improvement over IPSEC/IKE PFS transmission, which was what Soul-Burn666 was originally talking about.
And if you think "Eve" wouldn't keep working at a Sufficiently Important message for decades, then you have not studied enough history.
The ONLY thing which is outright impossible is One Time Pad
Well, yes, which is the point: Quantum encryption is a one time pad, furthermore with absolutely guaranteed security in pad generation and distribution. There are several possible non-algorithmic weaknesses to an ordinary one-time pad:
* Alice must make a truly random pad.
* The pad must not be intercepted and copied by Eve when Alice attempts to securely send it to Bob.
* The pad must NEVER be reused.
The laws of quantum mechanics insure that the QE pad is random and non-reusable, and that any interception of a QE message precludes the message from being transmitted, while alerting Alice and Bob to Eve's presence.
it's not truly practical, unless...
EXACTLY! It's that "unless" that some people are worried about. Some secrets need to be kept secure against even 50 years or more of advancing technology. Quantum encryption seems to be the trump card, taking code breaking that final step from the impractical to the impossible. The only attack remaining is interception of the plain text that exists at either end-- a weakness of all encryption methods that do not use Write-Only Memory storage. =)
So, how does the hazard of the dust from the two computers and eight laptops in my office compare to the hazard of the fumes from the new carpet and fresh paint that my boss insisted on?
Satan! Satan! Introibo ad altare dei nostri Satani!!
Ahem. Anyway....
I never said "good" products, I said "usable"-- try a dictionary. That Microsoft has made it's money essentially by selling a gold plated turd at gunpoint is (oddly) irrelevant to the article's discussion of an even more basic problem looming at Microsoft, to wit:
"The company is addicted to the revenue from these flagship products and is afraid to go in new directions that might initially hurt the bottom line."
They can't get bigger in the markets they're in, they're facing growing competition in those markets, and they're either unwilling or too incompetent to go after new markets. Oh, and now everyone hates their guts. Doomed, doomed, doomed.
If the combination to a safe is lost or unknown, it still can be opened. It may require a locksmith, several hours of trying combinations brute force, or just plain drilling the safe, but it is possible.
Time locks, I suppose, would make non-destructive methods harder; but if it's important enough, and you can drag in heavy enough equipment, you can drill even a bank vault safe open. Which, by the way, is NOT a random example: one consulting company I worked for was based in a remodeled bank building... which was the root cause of an "unfortunate incident".
Easy enough on recent Macs; boot to target mode, and the Mac's an external FW drive. (NB: this can be disabled by owner.)
Much Unix/Windows stuff gives way to a Knoppix-type boot CD-- about the easiest "parallel installation" possible.
My standard computer B&E tools:
Knoppix Linux-on-CD distro
Two USB/FW drive enclosures with cables (a SCSI enclosure, and USB/SCSI adapter for it, are in my advanced kit).
One 1GB ATA Hard drive, with DOS and a general Clear-CMOS utility. (SCSI version is in my advanced kit.)
Offline NT Password editor floppy.
DOS/Clear-CMOS boot floppy.
One "friggin huge" hard drive for putting retrieved data onto. (The first 5GB is a HFS partition with Mac OSX.3, followed by a 32GB FAT32 partition, with the remaining couple hundred GB also formatted FAT32.)
Screwdriver (Philips/Standard reversible combo)
The advanced kit also includes dual boot Windows/Linux and OS X/Debian laptops; a USB/FW DVD drive; Windows, Mac OS 9&X, Linux, and Solaris-x86 install disks; crossover ethernet and serial cables; a Torx driver set; lockpicks, bolt cutters, a mini-sledge, and a 1-liter flask for the liquid helium-- which needs to be filled shortly before using. (Haven't needed that yet, though.)
Various combinations of these will retrieve from almost anything... but be wary of RAID arrays and encrypted (eg: Windows EFS) folders; inexpert attempts may make the data unretrievable.
This guy can say anything he wants, but it won't change the fact that MS is *definitely* doing things 'right'.
Almost-- and thus, you miss the point of what he is saying. "Microsoft has *definitely* done things 'right'" would be more accurate.
With Windows 95, it created an operating system usable by the masses, with new features that everyone really wanted to upgrade to-- Internet Access. Windows 98 added improved driver support, particularly for USB. Windows ME added diddly-squat... and it's sales were mediocre. Windows 2000 turned the NT branch into an almost-consumer usable product; Windows XP put a pretty coat of frosting on that, and marginally improved stability and usability.
From my understanding of the history of technology, the Windows OS has been paralleling the development of every other technological tool in history, software or otherwise. You come up with an idea for something to do a job; you get it into a marginally workable form, and people try it; you improve it, and if you get lucky and it's useful enough, eveyone beats a path to your door. You may even make a few more "new and improved" versions. But eventually, you have a mature piece of technology, like egrep, or the pocket knife.
And demand peaks-- because a lot of people HAVE one already, thank you, I'll use it until it wears out. Oh, there's a new Swiss army knife with Torx bits? Maybe I'll look into that when my current knife breaks.
Windows (mostly) works. What the bulk of the masses want to do, it can let them do. It could be more stable, but that's something people feel they should get for free with their CURRENT version-- making people pay for that is tricky.
Since the year September Never Ended, the number of people who want to have a computer has been on the rise. Multi-computer households aren't uncommon. But the number of new purchases is peaking-- and the second computer in the house is often a hand-me-down.
Microsoft is at a point where there isn't much more obvious "new and improved" to put on for the consumer, with both their Office and OS-- so upgrade sales will fall off. Instead of people upgrading OS every two to three years, they'll upgrade every five to nine-- by buying a new computer after the old one dies. Of course, M$ could stop supporting the older software... with bad consequences for (in turn) security for those machines using the software, performance for those networks connected to those machines, and network-dependent software performance for any current Windows machines connected to the network. Ooops.
The article isn't suggesting M$ will go away. What it does imply is that there may be a massive correction at some point in the not-too-distant future (I'd guess 5-10 years, but that's just me) that will cost it a large chunk (I'd guess ~65%?) of its current revenue stream and stock value, and that the measures it is trying now to protect its current revenue stream will make it more difficult to adapt to those leaner times.
(Of course, Apple is in danger of this trap, too. With the OS X.2, X.3, and now X.4 upgrades, it seems to be getting hooked on the upgrade revenue stream, and I'm not convinced users will remain enthusiasic. X.3 added two features of substance that my Mac users noticed and drooled over: Expose, and the return of color-coded files and folders. After seeing the price, of ten machines, two were upgraded for this.)
While gasoline and diesel are obvious and major uses for oil, they're not the only ones. Could this stuff be used as a feed for plastics production? At what increase or decrease in the cost/quality of the plastics?
One of them is that just as it is with oil, uranium, too, will eventually run out. It is not a renewable source of energy like the algae would be.
We've a much better timetable on Uranium than Oil, though, even without the breeder reactor option (which puts the Thorium supply into play). And oceanic extraction is not out of the question; while not presently economical, it's been done, and demand would encourage further research to improve the technology. (It would also do bad things to the price of gold, but that's another story.)
Not that I'm entirely disagreeing, mind you. But one of the main things software helps you make is more software... which (DeToqueville types claim) will go to zero value as a result of the complete unprotection of IP. Talented people may be able to develop and refine operating systems as a hobby these days, but they have to earn a living before they put effort into their hobbies. And increases in productivity mean nothing if what you produce becomes worthless.
The patent and copyright system in the US was created to try to balance the need to give creative types some rewards for their efforts (to encourage progress and new thought) while enabling society to reap the benfits at limited cost and encouraging sucessive development. Recent legislation has begun badly imbalancing this towards benefiting creators-- or worse, their descendants. Weak examiniations by the patent office exacerbate the problem.
Bill Gates' basic point from 1976 was that, if you do good work, you should be able to get paid for it. And, from an economics standpoint, more people tend to be inclined to do that work if they get rewarded.
Of course, at the time writing software was a highly arcane and rare skill. These days, Microsoft's business is becoming more and more like prostitution in a college town: hard to make a living at because so many talented amateurs are giving a comparable product away for free.
Software to do a job appears to pass through three stages: where nobody knows how to do it, where an oligarchy knows how to do it, and where nearly every shmuck knows how to do it. As time progresses, and computer skills have spread, more and more things move from the first category to the second, and then the second to the third.
But you can only make a boodle of cash if what you're doing is in the middle category. What scares Bill is that almost all of Microsofts gigabucks of revenue come from Operating Systems and Office Suites... and Linux and Open Office have started moving (via the GPL) both of those from the hands of the oligarchy to the hands of the masses.
The DeToqueville people are whining about this trickle down trend as the third part of their "three edged sword". In this, they are unfortunately like King Canute and the tide. The solution, obviously, is to be move more things from what nobody can do into the hands of the oligarchy. Of course, this means that those (like Microsoft) cannot rest on their Intellectual laurel Property, but must keep working hard with no assurance they will be the oligarchs who get the next amazing idea... as Google seems to have demonstrated. It may well be that operating systems and office suites will not be where the smart people make their money in the future, but on organizing these tools to make work go smoother (like IBM does). Of course, to make money this way (for long), your CLIENT has to be making money producing something-- which, if IP becomes worthless, won't be an information economy product?
The DeToqueville institute may have some point with the first edge of their sword (as bad as that metaphor becomes), in that the GPL may be TOO STRONG a protection to encourage inventors properly... which I will suggest as a student term paper topic, rather than blather on about here. =)
Their second edge I consider contemptible. Yes, giving away Linux is providing jump starts to lower-income countries. As a fat, lazy American, I find the disparity in the global distribution of wealth digusting, and if adjusting that can be done by giving the poor oportunities to become richer, I can accept that it means that the rich have to work harder to stay that way.
A more interesting point that they raise is the shortsightedness of outsourcing in the effect that it has on redistribution of intellectual power. I think this will be the biggest long-term threat of outsourcing-- the gutting of the American skill set by failing to train replacements for the baby boomers.
The alternative being rejected by "only" is that credibility could be earned quickly
That would be "you earn only over time", while Markworth said "you only earn over time". While I have no doubt you are correct in what he INTENDED to mean, that's not what he actually said. And, given how HP has been doing lately, I found it notably ironic.
You'd think more CS geeks would understand the importance of being exact in what they say. While the shifting the position of modifiers in English can be subtle, the difference in meaning from doing so can be as profound as between "p *= ++i" and "p *= i++".
Mac OS X supports the use of 2nd button for pulling up "context menus", similar to right clicking in Windows. Of course, it's the same as control-clicking with the single button mouse, but it is supported. OS X also supports the scrollwheel to some extent (behaving mostly as you'd expect), which PearPC doesn't yet.
I once asked an Apple engineer why you couldn't GET a two-button mouse when buying a new Apple. It was implied that someone "with a huge amount of control over the design process" was still adamantly opposed to the 2nd button. I wonder who....
"Credibility is something you only earn over time"
Nope. You also can lose it over time, which HP and Compaq are notorious for. At this point, I trust the HP mid-to-high-end laser printers... and nothing else they make. I used to swear by HP scanners and calculators, but almost everything they make is going downhill in ergonomics and durability, even when the performance isn't crap to begin with.
I think you mean "assume a spherical cow in a vacuum".
Er, no, not in this case. After you consider the special case of the infinite sized spherical cow, you assume a finite spherical cow enclosed in an infinite volume of some partially reflective material-- vacuum being a special case (where the reflectivity is zero) that falls out of the calculations afterwards.
Assuming your spherical cow in a vacuum gives a bomb design that's a little too rough-and-ready for my taste-- it tends goes off on the assembly table, not at the target site.
The most dangerous materials are the extremely "hot" ones that are fresh out of a reactor. In order to be that hot, they have a half-life of seconds to barely a few years.
Cesium-137 has a half life about 30 years; long enough to last, short enought for relatively small volumes to be quite hellishly radioactive-- about 80 curies per gram, if I recall.
In the end, you'll pretty much do nothing more than increase everyone's chance of getting cancer. <sarcasm>What an effective terrorist weapon</sarcasm>
You've obviously never been personally involved in political debates on locating nuclear facilities; a large fraction of the population has hysterical phobias about radioactivity, even when there is no real danger. (I've seen a ditz go into hysterics on learning her skeleton was mildly radioactive from the natural potassium.)
Furthermore, while there will be few, if any, people getting an LD50/60 dose from a radiation dust bomb, cleaning up such an irradiated area could be prohibitively expensive. Failure to clean it up would result in an highly non-trivial increase in cancer rates in the area-- enough to make five-pack-a-day smoking look perfectly safe.
There's also the question as to whether or not Bin Laden would have competent enough people to know what they're stealing. For example, spreading a bunch of plutonium (Alpha Emitter) would be laughable.
Have you even taken a radiation health physics class? Alpha emitters are quite dangerous under the right conditions-- as you yourself noted, the real danger is in inhaling or ingesting radioisotopes. Furthermore, when you deal with a radiation source that's internal, alphas are about the worst of the lot, due to the high absorbtion of the radiation over short distnaces. As I noted, this is why you powderize the radioisope beforehand for this sort of weapon: to increase the chance of dust particles being inhaled.
The threat from a radiologic dust bomb isn't the initial short term exposure; it's the long term threat.
Slashdot Mirror
Um, I really don't think we want lots of people able to develop biological weapons in their basement. We already have enough problems with script kiddies making computer viruses, you'd think they'd learn.
This may be one of those technologies which creates a problem, the resolution of which is that the civilization making it gets knocked back to where it can no longer make the technology. (Classic examples from Science Fiction include certain general-purpose teleporters, as discussed in Niven's classic "On the Theory and Practice of Teleportation", and to a lesser degree the time viewer in Asimov's "The Dead Past".) I suppose that's one solution to the Fermi Paradox....
Depends what platform and browser you use. Beanie-W, problem go bye-bye.
But definitely don't try it on Windows/IE.
[People] want to profit [...], but that they can't make a short sale any other way.
Umm... so, it's all right to make money by selling something you don't even own the rights to sell? Like, say, SCO selling the right to use Linux?
Radio Frequency interference is a bad thing in airplanes; while not ruling out trying any of those RF options, I'd be a little cautious in risking wireless in a space station if I'm going to be breathing on it.
True, the slashdot article link is to the torture droid and not the Training Remote, but in Dubya's brave new world, IT-0 may be what NASA had in mind after all-- gotta keep close watch on them astronauts. Shifty, the lot of 'em.
While you're likely right about the chances, this quote from Sun's COO would qualfy as "mentioned by Sun", methinks.
In an interview with IDG News Service last month, Schwartz said "maybe we'll GPL it" and "We're still looking at that".
[W]hy not have Sun, IBM, Red Hat, SuSe and whoever else is pissed at SCO get together and spend 78 million and buy the bastards?
It's a matter of principle: "If you will not set a good example, you shall serve as a terrible warning." Or, more specifically here, IBM's version of it: "If you sue IBM, we will destroy you." Doing so serves to discourage "sue Big Blue" as an exit strategy for other failing companies-- a corollary of the old principle about Danegeld. Besides, IBM has most of these lawyers on salary... it's good to keep them in practice, just like it's good to keep your knives sharpened. =)
Unsubstatiated rumor has it that members of IBM's legal team have been informed that they will be considered to have failed if SCO even exists after this lawsuit ends.
Read more Groklaw.
The article parent links to discusses a separate motion to delay depositions--which, yes, was denied-- not delay the overall trial schedule. The motion to bifurcate the trial, and the motion to extend the Discovery/Deposition period before trial, are reported as being taken under advisement.
Yes, yes, quantum encryption precludes interception; ergo, unlike with IPSEC, "Eve" can't duplicate the QE message during its transfer, store the encoded message for 50 years, and then crack the code with Any Sufficiently Advanced Technological Improvement. So yes, it's useless for protecting storage-- as I noted, the plaintext on either end is still vulnerable-- but it does provide an improvement over IPSEC/IKE PFS transmission, which was what Soul-Burn666 was originally talking about.
And if you think "Eve" wouldn't keep working at a Sufficiently Important message for decades, then you have not studied enough history.
The ONLY thing which is outright impossible is One Time Pad
Well, yes, which is the point: Quantum encryption is a one time pad, furthermore with absolutely guaranteed security in pad generation and distribution. There are several possible non-algorithmic weaknesses to an ordinary one-time pad:
* Alice must make a truly random pad.
* The pad must not be intercepted and copied by Eve when Alice attempts to securely send it to Bob.
* The pad must NEVER be reused.
The laws of quantum mechanics insure that the QE pad is random and non-reusable, and that any interception of a QE message precludes the message from being transmitted, while alerting Alice and Bob to Eve's presence.
it's not truly practical, unless...
EXACTLY! It's that "unless" that some people are worried about. Some secrets need to be kept secure against even 50 years or more of advancing technology. Quantum encryption seems to be the trump card, taking code breaking that final step from the impractical to the impossible. The only attack remaining is interception of the plain text that exists at either end-- a weakness of all encryption methods that do not use Write-Only Memory storage. =)
So, how does the hazard of the dust from the two computers and eight laptops in my office compare to the hazard of the fumes from the new carpet and fresh paint that my boss insisted on?
Yeah, I kinda thought so.
If I recon correctly, gasoline fumes contains beneze. Have we stopped using gas ?
No, but most western countries have put limits on the amount of benzene permissible in gasoline, eg: USA, Canada, etc.
I have recently learnt that IPSEC/IKE does indeed give you PFS, perfect forward secrecy.
Tsk, tsk. Even that only uses a 1024 bit key, so I only need to try 1.8e+308 or so possible keys to find the right one-- not currently practical, but a few years of Moore's law might render the problem solvable within the lifetime of the known universe, even precluding a major breakthrough in quantum computing.
There's a difference between problems that are absurdly difficult, and problems that are outright impossible.
You're Satan, aren't you.
Satan! Satan! Introibo ad altare dei nostri Satani!!
Ahem. Anyway....
I never said "good" products, I said "usable"-- try a dictionary. That Microsoft has made it's money essentially by selling a gold plated turd at gunpoint is (oddly) irrelevant to the article's discussion of an even more basic problem looming at Microsoft, to wit:
"The company is addicted to the revenue from these flagship products and is afraid to go in new directions that might initially hurt the bottom line."
They can't get bigger in the markets they're in, they're facing growing competition in those markets, and they're either unwilling or too incompetent to go after new markets. Oh, and now everyone hates their guts. Doomed, doomed, doomed.
If the combination to a safe is lost or unknown, it still can be opened. It may require a locksmith, several hours of trying combinations brute force, or just plain drilling the safe, but it is possible.
Time locks, I suppose, would make non-destructive methods harder; but if it's important enough, and you can drag in heavy enough equipment, you can drill even a bank vault safe open. Which, by the way, is NOT a random example: one consulting company I worked for was based in a remodeled bank building... which was the root cause of an "unfortunate incident".
Easy enough on recent Macs; boot to target mode, and the Mac's an external FW drive. (NB: this can be disabled by owner.)
Much Unix/Windows stuff gives way to a Knoppix-type boot CD-- about the easiest "parallel installation" possible.
My standard computer B&E tools:
Knoppix Linux-on-CD distro
Two USB/FW drive enclosures with cables (a SCSI enclosure, and USB/SCSI adapter for it, are in my advanced kit).
One 1GB ATA Hard drive, with DOS and a general Clear-CMOS utility. (SCSI version is in my advanced kit.)
Offline NT Password editor floppy.
DOS/Clear-CMOS boot floppy.
One "friggin huge" hard drive for putting retrieved data onto. (The first 5GB is a HFS partition with Mac OSX.3, followed by a 32GB FAT32 partition, with the remaining couple hundred GB also formatted FAT32.)
Screwdriver (Philips/Standard reversible combo)
The advanced kit also includes dual boot Windows/Linux and OS X/Debian laptops; a USB/FW DVD drive; Windows, Mac OS 9&X, Linux, and Solaris-x86 install disks; crossover ethernet and serial cables; a Torx driver set; lockpicks, bolt cutters, a mini-sledge, and a 1-liter flask for the liquid helium-- which needs to be filled shortly before using. (Haven't needed that yet, though.)
Various combinations of these will retrieve from almost anything... but be wary of RAID arrays and encrypted (eg: Windows EFS) folders; inexpert attempts may make the data unretrievable.
This guy can say anything he wants, but it won't change the fact that MS is *definitely* doing things 'right'.
Almost-- and thus, you miss the point of what he is saying. "Microsoft has *definitely* done things 'right'" would be more accurate.
With Windows 95, it created an operating system usable by the masses, with new features that everyone really wanted to upgrade to-- Internet Access. Windows 98 added improved driver support, particularly for USB. Windows ME added diddly-squat... and it's sales were mediocre. Windows 2000 turned the NT branch into an almost-consumer usable product; Windows XP put a pretty coat of frosting on that, and marginally improved stability and usability.
From my understanding of the history of technology, the Windows OS has been paralleling the development of every other technological tool in history, software or otherwise. You come up with an idea for something to do a job; you get it into a marginally workable form, and people try it; you improve it, and if you get lucky and it's useful enough, eveyone beats a path to your door. You may even make a few more "new and improved" versions. But eventually, you have a mature piece of technology, like egrep, or the pocket knife.
And demand peaks-- because a lot of people HAVE one already, thank you, I'll use it until it wears out. Oh, there's a new Swiss army knife with Torx bits? Maybe I'll look into that when my current knife breaks.
Windows (mostly) works. What the bulk of the masses want to do, it can let them do. It could be more stable, but that's something people feel they should get for free with their CURRENT version-- making people pay for that is tricky.
Since the year September Never Ended, the number of people who want to have a computer has been on the rise. Multi-computer households aren't uncommon. But the number of new purchases is peaking-- and the second computer in the house is often a hand-me-down.
Microsoft is at a point where there isn't much more obvious "new and improved" to put on for the consumer, with both their Office and OS-- so upgrade sales will fall off. Instead of people upgrading OS every two to three years, they'll upgrade every five to nine-- by buying a new computer after the old one dies. Of course, M$ could stop supporting the older software... with bad consequences for (in turn) security for those machines using the software, performance for those networks connected to those machines, and network-dependent software performance for any current Windows machines connected to the network. Ooops.
The article isn't suggesting M$ will go away. What it does imply is that there may be a massive correction at some point in the not-too-distant future (I'd guess 5-10 years, but that's just me) that will cost it a large chunk (I'd guess ~65%?) of its current revenue stream and stock value, and that the measures it is trying now to protect its current revenue stream will make it more difficult to adapt to those leaner times.
(Of course, Apple is in danger of this trap, too. With the OS X.2, X.3, and now X.4 upgrades, it seems to be getting hooked on the upgrade revenue stream, and I'm not convinced users will remain enthusiasic. X.3 added two features of substance that my Mac users noticed and drooled over: Expose, and the return of color-coded files and folders. After seeing the price, of ten machines, two were upgraded for this.)
While gasoline and diesel are obvious and major uses for oil, they're not the only ones. Could this stuff be used as a feed for plastics production? At what increase or decrease in the cost/quality of the plastics?
One of them is that just as it is with oil, uranium, too, will eventually run out. It is not a renewable source of energy like the algae would be.
We've a much better timetable on Uranium than Oil, though, even without the breeder reactor option (which puts the Thorium supply into play). And oceanic extraction is not out of the question; while not presently economical, it's been done, and demand would encourage further research to improve the technology. (It would also do bad things to the price of gold, but that's another story.)
Not that I'm entirely disagreeing, mind you. But one of the main things software helps you make is more software... which (DeToqueville types claim) will go to zero value as a result of the complete unprotection of IP. Talented people may be able to develop and refine operating systems as a hobby these days, but they have to earn a living before they put effort into their hobbies. And increases in productivity mean nothing if what you produce becomes worthless.
The patent and copyright system in the US was created to try to balance the need to give creative types some rewards for their efforts (to encourage progress and new thought) while enabling society to reap the benfits at limited cost and encouraging sucessive development. Recent legislation has begun badly imbalancing this towards benefiting creators-- or worse, their descendants. Weak examiniations by the patent office exacerbate the problem.
Bill Gates' basic point from 1976 was that, if you do good work, you should be able to get paid for it. And, from an economics standpoint, more people tend to be inclined to do that work if they get rewarded.
Of course, at the time writing software was a highly arcane and rare skill. These days, Microsoft's business is becoming more and more like prostitution in a college town: hard to make a living at because so many talented amateurs are giving a comparable product away for free.
Software to do a job appears to pass through three stages: where nobody knows how to do it, where an oligarchy knows how to do it, and where nearly every shmuck knows how to do it. As time progresses, and computer skills have spread, more and more things move from the first category to the second, and then the second to the third.
But you can only make a boodle of cash if what you're doing is in the middle category. What scares Bill is that almost all of Microsofts gigabucks of revenue come from Operating Systems and Office Suites... and Linux and Open Office have started moving (via the GPL) both of those from the hands of the oligarchy to the hands of the masses.
The DeToqueville people are whining about this trickle down trend as the third part of their "three edged sword". In this, they are unfortunately like King Canute and the tide. The solution, obviously, is to be move more things from what nobody can do into the hands of the oligarchy. Of course, this means that those (like Microsoft) cannot rest on their Intellectual laurel Property, but must keep working hard with no assurance they will be the oligarchs who get the next amazing idea... as Google seems to have demonstrated. It may well be that operating systems and office suites will not be where the smart people make their money in the future, but on organizing these tools to make work go smoother (like IBM does). Of course, to make money this way (for long), your CLIENT has to be making money producing something-- which, if IP becomes worthless, won't be an information economy product?
The DeToqueville institute may have some point with the first edge of their sword (as bad as that metaphor becomes), in that the GPL may be TOO STRONG a protection to encourage inventors properly... which I will suggest as a student term paper topic, rather than blather on about here. =)
Their second edge I consider contemptible. Yes, giving away Linux is providing jump starts to lower-income countries. As a fat, lazy American, I find the disparity in the global distribution of wealth digusting, and if adjusting that can be done by giving the poor oportunities to become richer, I can accept that it means that the rich have to work harder to stay that way.
A more interesting point that they raise is the shortsightedness of outsourcing in the effect that it has on redistribution of intellectual power. I think this will be the biggest long-term threat of outsourcing-- the gutting of the American skill set by failing to train replacements for the baby boomers.
The alternative being rejected by "only" is that credibility could be earned quickly
That would be "you earn only over time", while Markworth said "you only earn over time". While I have no doubt you are correct in what he INTENDED to mean, that's not what he actually said. And, given how HP has been doing lately, I found it notably ironic.
You'd think more CS geeks would understand the importance of being exact in what they say. While the shifting the position of modifiers in English can be subtle, the difference in meaning from doing so can be as profound as between "p *= ++i" and "p *= i++".
Mac OS X supports the use of 2nd button for pulling up "context menus", similar to right clicking in Windows. Of course, it's the same as control-clicking with the single button mouse, but it is supported. OS X also supports the scrollwheel to some extent (behaving mostly as you'd expect), which PearPC doesn't yet.
I once asked an Apple engineer why you couldn't GET a two-button mouse when buying a new Apple. It was implied that someone "with a huge amount of control over the design process" was still adamantly opposed to the 2nd button. I wonder who....
"Credibility is something you only earn over time"
Nope. You also can lose it over time, which HP and Compaq are notorious for. At this point, I trust the HP mid-to-high-end laser printers... and nothing else they make. I used to swear by HP scanners and calculators, but almost everything they make is going downhill in ergonomics and durability, even when the performance isn't crap to begin with.
I think you mean "assume a spherical cow in a vacuum".
Er, no, not in this case. After you consider the special case of the infinite sized spherical cow, you assume a finite spherical cow enclosed in an infinite volume of some partially reflective material-- vacuum being a special case (where the reflectivity is zero) that falls out of the calculations afterwards.
Assuming your spherical cow in a vacuum gives a bomb design that's a little too rough-and-ready for my taste-- it tends goes off on the assembly table, not at the target site.
The most dangerous materials are the extremely "hot" ones that are fresh out of a reactor. In order to be that hot, they have a half-life of seconds to barely a few years.
Cesium-137 has a half life about 30 years; long enough to last, short enought for relatively small volumes to be quite hellishly radioactive-- about 80 curies per gram, if I recall.
In the end, you'll pretty much do nothing more than increase everyone's chance of getting cancer.
<sarcasm>What an effective terrorist weapon</sarcasm>
You've obviously never been personally involved in political debates on locating nuclear facilities; a large fraction of the population has hysterical phobias about radioactivity, even when there is no real danger. (I've seen a ditz go into hysterics on learning her skeleton was mildly radioactive from the natural potassium.)
Furthermore, while there will be few, if any, people getting an LD50/60 dose from a radiation dust bomb, cleaning up such an irradiated area could be prohibitively expensive. Failure to clean it up would result in an highly non-trivial increase in cancer rates in the area-- enough to make five-pack-a-day smoking look perfectly safe.
There's also the question as to whether or not Bin Laden would have competent enough people to know what they're stealing. For example, spreading a bunch of plutonium (Alpha Emitter) would be laughable.
Have you even taken a radiation health physics class? Alpha emitters are quite dangerous under the right conditions-- as you yourself noted, the real danger is in inhaling or ingesting radioisotopes. Furthermore, when you deal with a radiation source that's internal, alphas are about the worst of the lot, due to the high absorbtion of the radiation over short distnaces. As I noted, this is why you powderize the radioisope beforehand for this sort of weapon: to increase the chance of dust particles being inhaled.
The threat from a radiologic dust bomb isn't the initial short term exposure; it's the long term threat.