2 AUD per game is typical for most up-to-the-minute games - for something *really* special you can occasionally pay 3 AUD. Of course with the Pacific Peso the way it is at the moment that's less than a pound anyway . . .
Dunno about PARC itself, but Xerox's research division were publishing quite a few interesting papers on HCI a couple of years ago. One paper was a discussion on why paper is still better for some tasks than a computer screen, which went considerably beyond the "you can take it with you and the print is clearer".
Any F1 racer that tried to use traction control would get lapped by the 3rd lap.
Wrong, wrong, and wrong. A racer that used traction control as configured for a road car would get lapped by the third lap. A racer who used a properly tuned traction-control system would lap considerably faster. Go check your F1 history. They weren't banned because they made drivers go slower . . ..:)
I'm sure you know that most professional drivers turn off things like traction control when doing hot laps, as they _can_ outperform the computer.
You, presumably, are also aware that road car ABS and traction control devices are going to be biased towards "safety" rather than getting the absolute most out of the car.
You should also know that this kind of gear was allowed in F1 up until about 1991 - the Lotus and Williams of that era had traction control, ABS, active suspension - frankly, from a technical point of view, those cars were far more interesting than the F1 cars of today. Why was it banned? For sure it wasn't banned because it made the cars slower. . . BTW, rumours persist that Ferrari has figured out a way to make the engine management system act as a traction control device on their current F1 car. I wonder why they'd do that?
For the record, the 500cc GP bikes also use a traction control system on wet days.
Do you allow a differential GPS? What about a gyro compass? If you have those two items the problem just got *much* easier. If you don't allow them, why not?
Personally, the task of "driving an F1 car" is so much down to definitions of "driving" to make the problem not very appealing.
Though I sure Frank Williams and Patrick Head would sign it up as soon as it was available.
And I'm sure Bernie Eccelstone would be trying to figure out how to get a cut of the robot's salary:)
If these guys are serious (and their web page makes me doubt it) you'd try a less expensive and easier racing class, such as Formula Ford, *first*, and once the technology was regularly outperforming the best FF drivers, then think about tackling F1.
The nice thing about Formula ford is that there are large numbers of near-identical cars available off-the-shelf for a realistic pricetag, yet the skills required to drive a formula ford well are very, very similar to an F1 car.
Have you heard of traction control, anti-lock braking, and the like? If these things didn't do their job better than the best drivers in the world could, they wouldn't have been banned from F1.
I agree that driving a race car fast is a complex and difficult skill (a couple of laps in a go-kart can show you that), but it doesn't mean that computers can't be programmed to do most of them, particularly those that don't involve other drivers.
It's not clear from the web site if this robotic car is to actually compete in real race conditions, or if they plan some farce where it's just doing speed laps solo?
I very much doubt whether any driver would be prepared to compete with a robotic car under race conditions. Would you?
My former department traditionally used Solaris, complete with the usual set of gnu tools installed, but Linux (and the odd *BSD) boxes were popping up on the desk of every grad student and teacher who got funding (a depressingly scarce commodity at Australian universities, unfortunately). Cash-strapped universities who are already familiar with Unix didn't need much persuading.
I'm just curious - is there any CS department out there left that *isn't* using at least some Linux or BSD boxes?
I know you're being funny, but I'm sure that the TLAs have as complete of an archive of USENET, public mailing lists, web sites and the like as they can possibly get. For a lot of interesting (to a government) topics you could get all the intelligence you'd ever want with some clever search techniques and decent analysts.
I doubt we'll ever have to worry about this - now that it's no longer emitting radiation it'll be virtually impossible to find, and over the next decades and centuries the patch of space we'd have to search to find them will grow bigger and bigger. Additionally, while predicting timelines for this kind of thing is a joke, I think a "warp drive" or even something that could accelerate, stop, and turn around to go back to earth at fast enough for the crew to not be dead is still at least a century away.
Combining those two factors, by the time we have the capability to retrieve an object from deep space, the amount of deep space we'll have to search to find them will be just too big, unless we get really exotic and have technologies like endless quantities of nanospaceships with surveillance gear, perhaps.
Could somebody with more background in this stuff please comment?
Well, I'm sure Germans and most Europeans do. Besides, I think what they're doing is good, not doing anything over Echelon is accepting it. Even if this lawsuit might not actually do anything, it might wake up a few people with ties.
And if significant evidence becomes public that the US is stealing trade secrets for commercial advantage, expect that fact to be used as a negotiating point in the next GATT (General Agreement on Tariffs and Trade) rounds.
Fucking Mars babies. We don't have the expertise to send people to Mars in a worthwhile manner yet. Fuck whatever pop-sci books you've read.
Well why aren't we doing them instead of diddling around repeating Skylab and Mir? BOOOORIIIING!
We are not in a publicity race with another superpower anymore.
And if "we" weren't, satellites would still probably be a theoretical possibility.
Another hail mary program like Apollo just to satisfy infants like you? I don't think so.
Yeah, another hail mary like Apollo would be kind of good. Or, alternatively, another 50 planetary probes and satellite observatories, which we could probably do for the same cost as the ISS and would have immeasurably more scientific as well as pissing-contest value.
Leave the development of space to the grownups.
Thank you for your holier-than-thou attitude. Perhaps, by the time I depart this earth, you grownups might just contemplate that anything beyond what we did in 1969 might be possible "in the forseeable future". Just don't justify an engineering white elephant by arguing that it's great to get experience in building engineering white elephants.
Yeah, but you don't do a project because the spinoffs are cool. You do it because the project is worthwhile in the first place. We'd get just as many spinoff technologies from working towards a manned Mars mission, and it's something actually worth doing as distinct from the ISS, which will likely produce stuff-all of interest.
The nice thing about internet gambling is that, unlike casino gambling which are totally anonymous and there are *no* checks on how much an individual can wager, it's quite possible to lock problem gamblers out of net betting, and impose limits on the amount they can lose. Why? Because you need *a credit card*, which can be used to track *all* the gambling you do. While it's not a perfect identifier, it's a heck of a lot better than real casinos can do.
But, of course, gambling and is Really Really Evil(TM), and the Internet is Really Really New and Scary (TM), so combining the two must be Armageddon in a lunch box.
It's been demonstrated again and again that for everything other than absolute audiophile gear (>$5000US systems, >$500 headphones) the limiting factor isn't the signal coming out of the CD player (unless it's a $50 special), it's the quality of the amplifiers, cables, speakers, and particularly the acoustics of the listening environment.
Get yourself a pair of really good speakers and amplifier and *then* worry about the quality of your CD player.
This man does not share the belief that most/.ers have, that being, the government in inherently evil and trying to screw us over. over.
Intelligence-gathering services and federal police agencies h
ave a long history of politicization and exceeding their authority. Perhaps you've forgotten what the FBI got up to under Hoover, or the CIA's activities in the 1980's?
Even in my home state of Victoria, Australia, an "intelligence bureau" was set up where police were infiltrated into various groups of political activists such as student groups, the Squatters' Union, and a whole bunch of mostly innocuous community organisations. While it might be arguably legitimate to surveil these groups (including reading their mail, tapping phone calls, and keeping extensive dossiers on people who *hadn't committed crimes*) if they are planning violent protests, most of them had no such plans, and indeed the officers performing the surveillance begged their superiors to stop wasting their time. In one particularly sad/amusing case, the officer concerned sympathised with the advocacy group to such an extent that he spent a substantial amount of time teaching the members of the group typing skills so they could advocate their views more effectively.
When a new government was elected and discovered the surveillance division, the minister concerned ordered that the surveillance be stopped and all files to be destroyed. However, the surveillance continued and files were retained for at least another five years in direct contravention of the wishes of their political bosses (who were *fully* empowered to direct the police in this way).
People who worry about "law and order", "national security" and "stability" go work for these organisations. People worried about "civil liberties" don't.
Oh sure, astronauts live in zero G for a few weeks. A few Russians went for a few months, but they come back to Earth on a stretcher because their muscles have turned to jello.
True, but apparently the story is a little more complex than that. Russian cosmonauts on Mir were supposed to do intensive exercise regimes to preserve muscle tone, but these exercise regimes failed to work. What instead happened was that the cosmonauts weren' actually *doing* their exercise routines - Shannen Lucid, the American astronaut who was up on Mir for ages, actually bothered to do her routines and was able to walk around virtually immediately after returning to Earth.
And while Mars is not zero G. It is roughly 1/3 G. Long term residence on Mars will weaken people, possibly to the point to where they can never return to Earth. Human lifespan on Mars may also be severely shortened.
It's equally possible that human lifespan will be substantially *lengthened*. Until we actually go and live there for a while, we won't know.
Quick definition for others: NP-hard means
every problem in NP can be transformed in polynomial time into an instance of this problem, but it doesn't necessarily mean that it's itself in NP. NP-complete = NP-hard + is in NP.
...solutions found in Quantum algorithms for simulating NP-hard problems in polynomial time.
Are you sure? I don't think that this is the case - while some algorithms for currently-intractable problems have been proposed for a theoretical quantum computer, none of these algorithms is for an NP-hard problem. References appreciated if I'm horribly wrong (which I may well be).
Re:This is an incorrect definition of NP
on
Does P = NP?
·
· Score: 2
You're quite - in fact, you've understated it - as if you just consider the single human being they're definitely time-limited, unlike a Turing machine, where for discussions of computatibility (as distinct from complexity, of course) time limits are irrelevant. I should have made that proviso more explicit.
Conversely, if you've read Roger Penrose, you'll know that he argues that human intelligence is non-computable.
While this is all a very interesting debate, the original point I was making still applies, I think - that how human thought relates to computability isn't really relevant to a discussion on P=/!=NP.
Thank you for your correction, though - it's nice to see that people with clue still read/. occasionally.
MODERATORS: Please mod the above post up
on
Does P = NP?
·
· Score: 1
If Mr Roca is correct (and I think he might be) the consequences of P=NP are far bigger than I think anyone else in this discussion has stated.
Whether he is correct or not is also worthy of discussion.
Re:NP Non-deterministic Polynomial
on
Does P = NP?
·
· Score: 2
Am I to understand that through this alrotithm can solve any NP problem in O(n^6)????
Not quite (and I apologise in advance for any details I skip over, or mistakes I've made, it's been a year or so since I got out of academia). To solve say, SAT (the satisfiability problem) using this algorithm, first you have put a converter over the top to convert SAT problems into instances of this problem. Say that given a SAT problem of size n, the converter takes O(n^2) time to do the conversion to the clique problem, and generates a clique problem of size O(n^2). Now, in this case the time the conversion takes isn't relevant, but the size of the generated clique problem is. The hybrid method will now take O((n^2)^6)) = O(n^8) time to solve the problem.
However, from a theoretical perspective, we don't really care whether it's O(n^2), O(n^8), or O(n^93), just as long as it's O(n^x) rather than O(y^n). In practice, I'd expect that if we ever found that P=NP through a O(n^50) algorithm, once the world's computer scientists and mathematicians got over their week-long party they'd quickly knock the polynomial factor down to something more reasonable.
I'm still betting that P!=NP though - even if it's just the "I can't do it, so it must be impossible" factor coming through:)
Re:This is an incorrect definition of NP
on
Does P = NP?
·
· Score: 3
Actually, bringing humans into this discussion is not particularly useful. We do not know whether humans are Turing-equivalent or not (humans may have abilities beyond a Turing machine), let alone whether they have the capabilities of a deterministic or non-deterministic Turing machine. They do have the abilities of a deterministic Turing machine (get a pen and paper and you can simulate the operation of one, so yes we do have at least the capabilities of a deterministic Turing machine).
Keep in mind also that "nondeterminism" in this context means much more than just "can't predict the system's behaviour". It has a much more exacting definition than that - to a first approximation imagine a program that can, whenever it needs to make a decision, executes a fork() call, and continues executing simultaneously down both paths, either one of which can find the answer.
Slashdot Mirror
2 AUD per game is typical for most up-to-the-minute games - for something *really* special you can occasionally pay 3 AUD. Of course with the Pacific Peso the way it is at the moment that's less than a pound anyway . . .
Dunno about PARC itself, but Xerox's research division were publishing quite a few interesting papers on HCI a couple of years ago. One paper was a discussion on why paper is still better for some tasks than a computer screen, which went considerably beyond the "you can take it with you and the print is clearer".
You, presumably, are also aware that road car ABS and traction control devices are going to be biased towards "safety" rather than getting the absolute most out of the car.
You should also know that this kind of gear was allowed in F1 up until about 1991 - the Lotus and Williams of that era had traction control, ABS, active suspension - frankly, from a technical point of view, those cars were far more interesting than the F1 cars of today. Why was it banned? For sure it wasn't banned because it made the cars slower. . . BTW, rumours persist that Ferrari has figured out a way to make the engine management system act as a traction control device on their current F1 car. I wonder why they'd do that?
For the record, the 500cc GP bikes also use a traction control system on wet days.
Personally, the task of "driving an F1 car" is so much down to definitions of "driving" to make the problem not very appealing.
And I'm sure Bernie Eccelstone would be trying to figure out how to get a cut of the robot's salary :)
The nice thing about Formula ford is that there are large numbers of near-identical cars available off-the-shelf for a realistic pricetag, yet the skills required to drive a formula ford well are very, very similar to an F1 car.
I agree that driving a race car fast is a complex and difficult skill (a couple of laps in a go-kart can show you that), but it doesn't mean that computers can't be programmed to do most of them, particularly those that don't involve other drivers.
I very much doubt whether any driver would be prepared to compete with a robotic car under race conditions. Would you?
I'm just curious - is there any CS department out there left that *isn't* using at least some Linux or BSD boxes?
I know you're being funny, but I'm sure that the TLAs have as complete of an archive of USENET, public mailing lists, web sites and the like as they can possibly get. For a lot of interesting (to a government) topics you could get all the intelligence you'd ever want with some clever search techniques and decent analysts.
Combining those two factors, by the time we have the capability to retrieve an object from deep space, the amount of deep space we'll have to search to find them will be just too big, unless we get really exotic and have technologies like endless quantities of nanospaceships with surveillance gear, perhaps.
Could somebody with more background in this stuff please comment?
And if significant evidence becomes public that the US is stealing trade secrets for commercial advantage, expect that fact to be used as a negotiating point in the next GATT (General Agreement on Tariffs and Trade) rounds.
Well why aren't we doing them instead of diddling around repeating Skylab and Mir? BOOOORIIIING!
And if "we" weren't, satellites would still probably be a theoretical possibility.
Yeah, another hail mary like Apollo would be kind of good. Or, alternatively, another 50 planetary probes and satellite observatories, which we could probably do for the same cost as the ISS and would have immeasurably more scientific as well as pissing-contest value.
Thank you for your holier-than-thou attitude. Perhaps, by the time I depart this earth, you grownups might just contemplate that anything beyond what we did in 1969 might be possible "in the forseeable future". Just don't justify an engineering white elephant by arguing that it's great to get experience in building engineering white elephants.
Yeah, but you don't do a project because the spinoffs are cool. You do it because the project is worthwhile in the first place. We'd get just as many spinoff technologies from working towards a manned Mars mission, and it's something actually worth doing as distinct from the ISS, which will likely produce stuff-all of interest.
But, of course, gambling and is Really Really Evil(TM), and the Internet is Really Really New and Scary (TM), so combining the two must be Armageddon in a lunch box.
Get yourself a pair of really good speakers and amplifier and *then* worry about the quality of your CD player.
That would be "Tomix", "Dickix", or "Harrix"...:)
Even in my home state of Victoria, Australia, an "intelligence bureau" was set up where police were infiltrated into various groups of political activists such as student groups, the Squatters' Union, and a whole bunch of mostly innocuous community organisations. While it might be arguably legitimate to surveil these groups (including reading their mail, tapping phone calls, and keeping extensive dossiers on people who *hadn't committed crimes*) if they are planning violent protests, most of them had no such plans, and indeed the officers performing the surveillance begged their superiors to stop wasting their time. In one particularly sad/amusing case, the officer concerned sympathised with the advocacy group to such an extent that he spent a substantial amount of time teaching the members of the group typing skills so they could advocate their views more effectively.
When a new government was elected and discovered the surveillance division, the minister concerned ordered that the surveillance be stopped and all files to be destroyed. However, the surveillance continued and files were retained for at least another five years in direct contravention of the wishes of their political bosses (who were *fully* empowered to direct the police in this way).
People who worry about "law and order", "national security" and "stability" go work for these organisations. People worried about "civil liberties" don't.
If there's money in it, without a doubt. And there's every possibility there will be money in it.
True, but apparently the story is a little more complex than that. Russian cosmonauts on Mir were supposed to do intensive exercise regimes to preserve muscle tone, but these exercise regimes failed to work. What instead happened was that the cosmonauts weren' actually *doing* their exercise routines - Shannen Lucid, the American astronaut who was up on Mir for ages, actually bothered to do her routines and was able to walk around virtually immediately after returning to Earth.
It's equally possible that human lifespan will be substantially *lengthened*. Until we actually go and live there for a while, we won't know.
Quick definition for others: NP-hard means every problem in NP can be transformed in polynomial time into an instance of this problem, but it doesn't necessarily mean that it's itself in NP. NP-complete = NP-hard + is in NP.
Are you sure? I don't think that this is the case - while some algorithms for currently-intractable problems have been proposed for a theoretical quantum computer, none of these algorithms is for an NP-hard problem. References appreciated if I'm horribly wrong (which I may well be).
Conversely, if you've read Roger Penrose, you'll know that he argues that human intelligence is non-computable.
While this is all a very interesting debate, the original point I was making still applies, I think - that how human thought relates to computability isn't really relevant to a discussion on P=/!=NP.
Thank you for your correction, though - it's nice to see that people with clue still read /. occasionally.
If Mr Roca is correct (and I think he might be) the consequences of P=NP are far bigger than I think anyone else in this discussion has stated.
Whether he is correct or not is also worthy of discussion.
Not quite (and I apologise in advance for any details I skip over, or mistakes I've made, it's been a year or so since I got out of academia). To solve say, SAT (the satisfiability problem) using this algorithm, first you have put a converter over the top to convert SAT problems into instances of this problem. Say that given a SAT problem of size n, the converter takes O(n^2) time to do the conversion to the clique problem, and generates a clique problem of size O(n^2). Now, in this case the time the conversion takes isn't relevant, but the size of the generated clique problem is. The hybrid method will now take O((n^2)^6)) = O(n^8) time to solve the problem.
However, from a theoretical perspective, we don't really care whether it's O(n^2), O(n^8), or O(n^93), just as long as it's O(n^x) rather than O(y^n). In practice, I'd expect that if we ever found that P=NP through a O(n^50) algorithm, once the world's computer scientists and mathematicians got over their week-long party they'd quickly knock the polynomial factor down to something more reasonable.
I'm still betting that P!=NP though - even if it's just the "I can't do it, so it must be impossible" factor coming through :)
Keep in mind also that "nondeterminism" in this context means much more than just "can't predict the system's behaviour". It has a much more exacting definition than that - to a first approximation imagine a program that can, whenever it needs to make a decision, executes a fork() call, and continues executing simultaneously down both paths, either one of which can find the answer.