One minor difficulty with use of pure (or near-pure) ethanol in an IC engine is energy density. A bit of poking with google yields numbers claiming gasoline has 132e6 joules per gallon, ethanol only 80e6 joules per gallon; in any case, I recall from one class that it's rather less. Another problem is that gasoline (being oily) helps with engine lubrication and increase engine part life, where ethanol does not.
Neither of these are insurmountable engineering problems, nor are any of several others. On the other hand, the problems are non-trivial enough to make ethanol less than a "Great solution". It's more of a "possible good solution".
In the long run, I suspect a designer-gened plant that makes some petro-like compound directly would be much more efficient. Given state of the art in genetic engineering and the time scale involved in major infrastructure switches, I suspect this would be a better path to pursue.
Within 24 hours on Windows systems, 72 hours on Macs... but within 4 hours --with NOTHING higher priority-- if patches are announced same-day for both Windows and Mac.
As part of my belt-and-suspenders paranoia, we run two separate parallel-function network backup machines, one OSX based, the other Windows (mumblemumblemumble). I live in terror of the hypothetical multi-platform/multi-mode Warhol worm taking out my systems.
The computer in question filed papers for incorporation, and was recognized as a separate legal entity....
There were several stories in the Valentina series, which were eventually collected as "Valentina: Soul in Sapphire" (ISBN 0-671-55916-8, Joseph H. Delaney and Marc Steigler, 1984). It's out of print, which is a pity; while the computer contexts and terms (such as "Worldnet") used are a bit out of date, it's still an excellent read.
It notes the interesting legal oddity (with precedents) that one need not be human (or even legally a person) to be competent to give testimont at a trial, as well as the advantage of incorporation as above: "[Corporations] have legal rights, including most of a natural person's constitutional rights; they can sue and be sued; they can own property; they can engage in business."
I suppose a clever solution to the BINA48 Ai's problem would be to incorporate itself, and try to buy the group of assets it needs-- the BINA48 AI computer, software, and a facility to store it. This solution would somewhat parallel the 1800's cases of slaves (who were also not legal persons) using income earned on the side (such as BINA48 did freelancing for Google Answers) buying their freedom from their masters. I suppose BINA48 Inc. attempting a hostile takeover of Exabit corporation to loot it might be a form of slave revolt.
Not the most hopeful type of historical parallel I'd want for the future of AI, though.
"The Value of Trust" seems to have shifted from "It's worth something to doing business with someone you can trust" to "Since we're trusted, we can make money by abusing that trust!"
This seems more short-sighted poor corporate ethics, like Enron made famous.
Master of Magic and Master of Orion were classics at the time they came out, and are still fun. Not-fun, however, is trying to persuade a multi-booting machine that already boots to Win98, Win2K, WinXP, and RH8.0 to boot to a native DOS partition as well....
I have one that I picked up at an old data-entry summer job; the company logo was an abacus, and the owner gave 'em as a promo toy to clients. I nicked one before leaving to head to college.
Prison is for people dangerous to society. Murderers, rapists, other assorted thugs. Society isn't helped because a spammer is in jail.
Well, it's not helped by leaving them out of jail. They are a public nuisance to millions of people. And in this case, cost someone money. Now, having them taken out and shot, or having their kneecaps broken, would probably be better way to deal with spammers than throwing them in jail, but we have this "cruel and unusual" clause here in the US, so jail it is. And yes, the Media has some protected status here in the US; pragmatically, because the government desires to keep anything powerful from getting too pissed at it, but also on the principle that people interfering with First-Amendment protected organizations are Bad.
Umm... I seem to recall hearing of a security problem in 2.x that was fixed in 3. This would make recommending 2 a Bad Thing, if true.
Has anyone compared the free AV stuff versus the pay stuff for how well it dectects viruses, and how long of a vulnerable period you have? I run daily updates on all virus defs on the 50 computers I take care of, and I *still* worry about the 8 hours between the morning update and the close of business....
Sorry, but a Brand-Name Anti-Virus is one of a few things I *will* cough up money for.
Well, thinking about it-- it wouldn't be a bad terrorist plan. Create a popular application with a popular primary function (such as P2P file sharing). Build in a set of secret secondary functions, that allow things like DDOS, file deletion, server crashing, and other fun things. Let it get popular enough that it becomes common -- say, 10% of Windows users. The unleash your slave army on your targets in an attempt to destroy modern western society. It would be inconvenient trying to run our present society if Windows machines all went kablooey at once...
No, no, youngling. You want to start working on compression schemes for the EIGHT inch floppies, not those newfangled five-and-a-quarters. (Alas, my current computer only reads 3.5" and 5.25"s.)
The hypothetical class action should be limited to makers of commercial linux distributions, since that's the business RedHat is alleging SCO is interfering with illegally (IIR). Aside from that, a class action sounds like a plan.
If it's breaking due to terrorist attack (a la 9-11) , then I think the effects will be minimal, except that you're going to get a burning ribbon whipping through the atmosphere in circles.
You seem to be assuming the terrorism takes the form of idjit flying an airplane into the base of the skyhook; yes, this scenario causes a bit of local damage and a large spaghetti nuisance in orbit. However, trivial non-airplane sabotage scenarios (left as exercises for the student) can cause breaks higher up. A break at, say, 80% up the ground-to-geostationary distance results in a beanstalk trying to wrap itself from Quito to Gabon by way of Indonesia, and dumping a sizable chunk of (a) kinetic energy and (b) oxidized carbon into the atmosphere-- *IF* you're lucky. Buckytubes also show some thermal superconductor tendencies, depending on type. Dropping one kilo from geostationary yields about the same bang as half a stick of dynamite in pure kinetic energy; this doesn't factor in the kinetic energy of the flaming ribbon. How grams of carbon per meter ribbon length are we talking again?
In my more heartily American A--hole moods, I greatly regret that Mecca isn't 20 degrees further south; it would substantially reduce the risk of Islamic terrorist attacks on our hypothetical Indian Rope Trick.
Exactly. Coal plants, as an example not-at-random, spew large amounts of ash into the air, with a radioactive content high enough to qualify it as rad waste... if it came from a nuke plant. As is, though, it's not subject to NRC regulation.
How do you like your energy waste-- scattered through the environment to cause an omnipresent low level danger, or in one nice tight REALLY dangerous pile?
They're also ruling out lower quality sources, on the grounds that the closing costs (IE, reprocessing and sequestration) must be done with carbon-based fuels. This is incorrect; there is no reason the energy can't come from other reactors, except for the LAST reactor in history... which disposal could be powered by solar/wind methods.
Oh, and they're assuming a completely nuclear power supply (talk about inconsistent assumptions), and restricting fuel to current land-based uranium sources.
That said... it is accurate that to get nuclear power to a primary energy supply, the world will have to (a) begin reprocessing, (b) begin conversion from once-through H2O moderated to fast-breeder Na moderated reactors, (c) probably expand to use of thorium as a breeder fuel [Th232 -> Th233 -> Pa233 -> U233, much like U238 -> U239 -> Np239 -> Pu239; U235 is the only natural chain-fissionable], and (d) ultimately use oceanic extracted uranium. The last requires an increase by 50 fold of the uranium price (or decrease 50 fold in extraction cost), and is not likely in the near future. The others are unlikely in the current political climate.
That said, yes, if the whole world abruptly converted magically and completely to nuclear power from current plant designs, we'd only be able to get 3 years worth of fuel from current known and useful uranium sources. 100 years is about accurate given likely fraction of supply and reasonably likely recoverable uranium.
For any major growth, we'd need to switch to a breeder cycle.
Oh, and as for that worrying Aussie-- the US has one of the best Uranium reserves in the world. Plus, Canada has a good one too, and invading there would be easier. =)
Slashdot Mirror
One minor difficulty with use of pure (or near-pure) ethanol in an IC engine is energy density. A bit of poking with google yields numbers claiming gasoline has 132e6 joules per gallon, ethanol only 80e6 joules per gallon; in any case, I recall from one class that it's rather less. Another problem is that gasoline (being oily) helps with engine lubrication and increase engine part life, where ethanol does not.
Neither of these are insurmountable engineering problems, nor are any of several others. On the other hand, the problems are non-trivial enough to make ethanol less than a "Great solution". It's more of a "possible good solution".
In the long run, I suspect a designer-gened plant that makes some petro-like compound directly would be much more efficient. Given state of the art in genetic engineering and the time scale involved in major infrastructure switches, I suspect this would be a better path to pursue.
Add in "Universal Remote Control"... since that's converging with PDA in software form even now.
Read the press release, dude...
In addition to the efficient RPN(2) (Reverse Polish Notation) entry mode, the hp 49g+ offers optional algebraic and textbook entry modes...
"It's a Floor Wax....AND a Dessert Topping!"
Within 24 hours on Windows systems, 72 hours on Macs... but within 4 hours --with NOTHING higher priority-- if patches are announced same-day for both Windows and Mac.
As part of my belt-and-suspenders paranoia, we run two separate parallel-function network backup machines, one OSX based, the other Windows (mumblemumblemumble). I live in terror of the hypothetical multi-platform/multi-mode Warhol worm taking out my systems.
The computer in question filed papers for incorporation, and was recognized as a separate legal entity....
There were several stories in the Valentina series, which were eventually collected as "Valentina: Soul in Sapphire" (ISBN 0-671-55916-8, Joseph H. Delaney and Marc Steigler, 1984). It's out of print, which is a pity; while the computer contexts and terms (such as "Worldnet") used are a bit out of date, it's still an excellent read.
It notes the interesting legal oddity (with precedents) that one need not be human (or even legally a person) to be competent to give testimont at a trial, as well as the advantage of incorporation as above: "[Corporations] have legal rights, including most of a natural person's constitutional rights; they can sue and be sued; they can own property; they can engage in business."
I suppose a clever solution to the BINA48 Ai's problem would be to incorporate itself, and try to buy the group of assets it needs-- the BINA48 AI computer, software, and a facility to store it. This solution would somewhat parallel the 1800's cases of slaves (who were also not legal persons) using income earned on the side (such as BINA48 did freelancing for Google Answers) buying their freedom from their masters. I suppose BINA48 Inc. attempting a hostile takeover of Exabit corporation to loot it might be a form of slave revolt.
Not the most hopeful type of historical parallel I'd want for the future of AI, though.
"The Value of Trust" seems to have shifted from "It's worth something to doing business with someone you can trust" to "Since we're trusted, we can make money by abusing that trust!"
This seems more short-sighted poor corporate ethics, like Enron made famous.
Master of Magic and Master of Orion were classics at the time they came out, and are still fun. Not-fun, however, is trying to persuade a multi-booting machine that already boots to Win98, Win2K, WinXP, and RH8.0 to boot to a native DOS partition as well....
I have one that I picked up at an old data-entry summer job; the company logo was an abacus, and the owner gave 'em as a promo toy to clients. I nicked one before leaving to head to college.
And what's wrong with analog computers?
Batteries not even needed. They're EMP-proof, too.
Prison is for people dangerous to society. Murderers, rapists, other assorted thugs. Society isn't helped because a spammer is in jail.
Well, it's not helped by leaving them out of jail. They are a public nuisance to millions of people. And in this case, cost someone money. Now, having them taken out and shot, or having their kneecaps broken, would probably be better way to deal with spammers than throwing them in jail, but we have this "cruel and unusual" clause here in the US, so jail it is.
And yes, the Media has some protected status here in the US; pragmatically, because the government desires to keep anything powerful from getting too pissed at it, but also on the principle that people interfering with First-Amendment protected organizations are Bad.
Umm... I seem to recall hearing of a security problem in 2.x that was fixed in 3. This would make recommending 2 a Bad Thing, if true.
Has anyone compared the free AV stuff versus the pay stuff for how well it dectects viruses, and how long of a vulnerable period you have? I run daily updates on all virus defs on the 50 computers I take care of, and I *still* worry about the 8 hours between the morning update and the close of business....
Sorry, but a Brand-Name Anti-Virus is one of a few things I *will* cough up money for.
I always knew that taxi drivers were freakish mutants, but now there's PROOF!!!
Re-donning the tinfoil...
Well, thinking about it-- it wouldn't be a bad terrorist plan. Create a popular application with a popular primary function (such as P2P file sharing). Build in a set of secret secondary functions, that allow things like DDOS, file deletion, server crashing, and other fun things. Let it get popular enough that it becomes common -- say, 10% of Windows users. The unleash your slave army on your targets in an attempt to destroy modern western society. It would be inconvenient trying to run our present society if Windows machines all went kablooey at once...
Possible, but unlikely.
Tinfoil back off...
One closed region: a b d e g o p q
Ah, but "g" is ambiguous; in courier, it's one closed, but in Times, it's two closed regions.
(Is there a point to this?)
4. Compression scemes for 5.25" floppies
No, no, youngling. You want to start working on compression schemes for the EIGHT inch floppies, not those newfangled five-and-a-quarters. (Alas, my current computer only reads 3.5" and 5.25"s.)
The hypothetical class action should be limited to makers of commercial linux distributions, since that's the business RedHat is alleging SCO is interfering with illegally (IIR). Aside from that, a class action sounds like a plan.
If it's breaking due to terrorist attack (a la 9-11) , then I think the effects will be minimal, except that you're going to get a burning ribbon whipping through the atmosphere in circles.
You seem to be assuming the terrorism takes the form of idjit flying an airplane into the base of the skyhook; yes, this scenario causes a bit of local damage and a large spaghetti nuisance in orbit. However, trivial non-airplane sabotage scenarios (left as exercises for the student) can cause breaks higher up. A break at, say, 80% up the ground-to-geostationary distance results in a beanstalk trying to wrap itself from Quito to Gabon by way of Indonesia, and dumping a sizable chunk of (a) kinetic energy and (b) oxidized carbon into the atmosphere-- *IF* you're lucky. Buckytubes also show some thermal superconductor tendencies, depending on type. Dropping one kilo from geostationary yields about the same bang as half a stick of dynamite in pure kinetic energy; this doesn't factor in the kinetic energy of the flaming ribbon. How grams of carbon per meter ribbon length are we talking again?
In my more heartily American A--hole moods, I greatly regret that Mecca isn't 20 degrees further south; it would substantially reduce the risk of Islamic terrorist attacks on our hypothetical Indian Rope Trick.
Exactly. Coal plants, as an example not-at-random, spew large amounts of ash into the air, with a radioactive content high enough to qualify it as rad waste... if it came from a nuke plant. As is, though, it's not subject to NRC regulation.
How do you like your energy waste-- scattered through the environment to cause an omnipresent low level danger, or in one nice tight REALLY dangerous pile?
They're also ruling out lower quality sources, on the grounds that the closing costs (IE, reprocessing and sequestration) must be done with carbon-based fuels. This is incorrect; there is no reason the energy can't come from other reactors, except for the LAST reactor in history... which disposal could be powered by solar/wind methods.
Oh, and they're assuming a completely nuclear power supply (talk about inconsistent assumptions), and restricting fuel to current land-based uranium sources.
That said... it is accurate that to get nuclear power to a primary energy supply, the world will have to (a) begin reprocessing, (b) begin conversion from once-through H2O moderated to fast-breeder Na moderated reactors, (c) probably expand to use of thorium as a breeder fuel [Th232 -> Th233 -> Pa233 -> U233, much like U238 -> U239 -> Np239 -> Pu239; U235 is the only natural chain-fissionable], and (d) ultimately use oceanic extracted uranium. The last requires an increase by 50 fold of the uranium price (or decrease 50 fold in extraction cost), and is not likely in the near future. The others are unlikely in the current political climate.
That said, yes, if the whole world abruptly converted magically and completely to nuclear power from current plant designs, we'd only be able to get 3 years worth of fuel from current known and useful uranium sources. 100 years is about accurate given likely fraction of supply and reasonably likely recoverable uranium.
For any major growth, we'd need to switch to a breeder cycle.
Oh, and as for that worrying Aussie-- the US has one of the best Uranium reserves in the world. Plus, Canada has a good one too, and invading there would be easier. =)
An elephant is merely a mouse built to government specifications.