I'd have to agree with you on optimizing. Chances are, though, that you could slim down further since not every page includes every feature of every library. The AJAX, CSS and other "enhancing" content never actually used on a given page never needs to be sent over with that page. "Page cleaning" before delivery should really be standard, IMHO -- what's the point of broadband if the page loads no faster than it did on a 56K modem?
(I've also seen far too many servers with page compression disabled because either the admin didn't know it could be enabled or because of bugs in the server.)
Of course, if Java hadn't had such limited communication with the browser page, the background libraries could be sent once for a site rather than once for a page, then cached locally on the client.
Precisely, and trying to shape human nature through regulation is where you run into the politicized nature of these environments. Yes, these systems should never be connected to the Internet (even by indirect means such as sneaker net) and yes fixing the problem must take human nature into account, but it must do so cleverly because applying the fix also involves human nature. That's a tough one.
To your example of someone stealing a car, I'd point out that yes the person who *stole* it may be guilty, but mistaken identity happens all the time and that means that you cannot go by one person's word alone that the person labelled as the thief is indeed the thief. There are all kinds of cases where vigilantes have attacked the wrong person for this very reason. Those vigilantes aren't part of the judicial system, so should they presume innocence? Yes. Obviously. Why obviously? Because the scientific method only works if you are trying to falsify things, it doesn't work if you are trying to prove things. (See Sherlock Holmes as to why this is important in criminal matters - he may be fictional, but he's a better role-model than Joe Thug.) If you start by assuming the person you think is guilty is, in fact, innocent then you can apply rigorous methods to falsify that hypothesis. Rigour is what matters.
To your example of AZ law, the Feds are investigating the sheriff in question (and seem on the verge of numerous indictments) because said sheriff DID regard that as sufficient grounds.
No, the police report only says what the police report says. Anyone can type out a document. Doesn't make the document true. Those of us from Britain are familiar with the Guildford 6 and the Birmingham 4, cases where false reports were made and falsified confessions (complete with forged signatures) were provided. We know that if these things could happen, they can happen and eventually will happen. You MUST therefore not accept any police report as being Absolute Truth, you must only accept it as being evidence offered that is no different from any other evidence, given no more or less weight, and scrutinized accordingly.
My own belief is that NO law, Federal or State, should be ignored, that ALL should be equal before the law, and that ALL evidence should be weighed and analyzed to the greatest practical* extent. Adversarial systems are something we're probably stuck with, and they serve a purpose, but monocultures are always defective and the current adversarial system in the US is definitely in the defective category.
*It would be nice if DNA evidence was analyzed by doing genomic decodes. Nice, but not practical. Maybe in 50 years. Perhaps. There will always be greater scrutiny technically possible, but you've got to draw some line or it gets absurd. In the case of DNA evidence, comparison of 8 markers isn't even remotely good enough - most labs wouldn't even break into a sweat if you upped the minimum to somewhere around 15-20. It would eliminate the false positives we're now seeing at no significant overhead in cost or time. That's an example of a practical threshold.
I seem to remember seeing SCADA vulnerabilities being added to vulnerability testing tools and IDS systems recently -- anyone know if this is related (ie: the tools now check for these non-existent flaws) or if the additions were to cover previously-reported bugs?
If the former, Siemens had best fix this damn fast. Infrastructure companies are in a corner - they don't have the cash for a major migration and alternative vendors are hardly thick on the ground. Some will be unable to afford decent security and others will be too politicized to secure their networks. Much of the infrastructure is too big and/or too expensive to duplicate, so the market is useless. The only place this can be fixed is at Siemens itself. The others that technically could won't and the rest can't.
The problem with the current paranoia over security is that you can't fix a fault you won't admit exists, companies won't deploy a fix if you tell them it's not needed, and so what you're ultimately left with is not security, merely obscurity.
Common sense is remarkably uncommon, but I guess you knew that as well. Pi is round, not square. And if I thought I was a guru, I'd be charging a hell of a lot more for my time. Spout off BS and I'll call you on your BS. If you don't want me calling you on it, don't spout it. It's very easy. You might even be able to achieve something like that.
Of course, since there are now thousands of tunnels, the border patrol is also going to need drone C-130 transport planes kitted with synthetic-aperture ground-penetrating radar. Which can be done, but is probably going to add a bit to the cost.
"Why should kids have all the fun? R/C toy for grown-ups includes all the accessories now found on the latest Real Action toys for children, in adult proportions. Real-action missiles come with realistic explosions."
1. There may well be people who have US citizenship (and therefore be entirely legal) who have no documentation to prove it (and thus be undocumented). I'm not going to pretend that this would be a large number, but if there is even one such person then they are an undocumented immigrant not an illegal immigrant. This would include US legitimate tourists who have been robbed, people born out of the country with at least one legitimate US parent, etc. The former can apply to the consulate, but that assumes they're rational. Rational people are an endangered species. The latter may or may not have access to the consulate, even if theory says they should have.
2. In the US (not sure about your country), a person is presumed innocent until proven guilty in a court of law. Well, that's the theory, at least. Legal doctrine therefore states these people -cannot- be "illegal immigrants" until proven such. In part, this is due to (1), but it's also because you don't want some mad Arizonian sheriff arresting anyone who "looks funny" and deporting them without lawful right to do so. And, no, saying someone "looks funny" is not a lawful right.
3. Once a person is at a detention centre, there is NO evidence of where they were arrested. There are corrupt police - hopefully not many, but it's definitely non-zero. It would not take much to take a lawful US citizen from within the US and make it appear like they're illegally there, especially if said citizen has no documentation on them (ie: they're undocumented). It's entirely plausible for police to eliminate homelessness by dumping the homeless over the border, and for hospitals to eliminate mental illness the same way. (You've seen the stories on hospitals dumping patients in skid row.) These would not be illegals, these would be undocumenteds.
I cannot tell from a police report or a media photo whether the person was legally entitled to be in the US. Nor can you. Nor can anyone. That is why we have courts. Judge without knowledge at your peril, because it is inevitable that when a society converts a potential for a crime into a crime in itself, you WILL be judged without knowledge yourself. And that's a path that goes downhill FAST.
(Most of those who are passionate about convicting without prosecuting would do well to remember that the road to hell is paved with "good intentions" -- not intentions that are actually good, merely intentions you can fool yourselves into believing are good.)
CC would be good because it's designed for multiple types of data.
I just checked OpenCores (the first of the hardware open source sites) and the preferred license is LGPL on there. I would say that if it's already in use for hardware specs and is working for that then it's a viable license to use.
This has been discussed before on Slashdot and we've even had people from the US Patent Office discuss what it is the patent office actually does. I am not to blame if you have the memory of a goldfish, that is your problem and you are welcome to it. No, I won't post links, there's a search tool and there's google. Use 'em. I'm tired of you youngsters wanting us old folk to do the work you're too lazy to do. Now gerroff my lawn!
The USPO doesn't review patents. It makes them available in case anyone wants to complain (there's a year window for that) but they leave it up to the courts to handle the legality of patents. The problem is, the courts defer to the USPO and assumes they've handled the legality of patents. If someone actually took responsibility for patents, there probably wouldn't be the current mess.
...multi-tasking hardware has been around for a while. Hell, I started off on one such device (the PET 3032 with 4040 disk drive was multi-tasking as the 4040 had an independent CPU, ROM and RAM, so you could feed it an instruction and it would independently go away to do stuff without tying up the main CPU). I see nothing particularly new or innovative in the concept.
I dunno if the Amulet CPU can multi-task at the hardware level (versus OS multi-tasking in software) but the modular nature of it means I would not be at all surprised.
I still think that plaintiffs should be able to feed CEOs of malignant companies to the salt water crocodiles. The TV coverage would be so much better than most of the regular programming and it shouldn't require that many before corporate practices clean right up.
That is something the courts have been hesitant to rule on but I believe that the US courts have ruled things like First Sale Doctorine, etc, take precedent over any such agreement. (ie: the EULA can't violate the law or otherwise impose a system contrary to the law.) Not being able to sue Sony in the event of Sony violating civil law - ie: denial of access to any system that can give relief - is usurping the courts entirely. This might not go down too well with judges, since if it's allowed, any product could have such a provision. If they allow one company to exempt itself from the legal system, they create a precedent (something judges are VERY loath to do) and case-law which would essentially state that any company could stipulate that a purchaser can't sue.
The civil court system depends heavily on people being able to sue each other. If the civil court system were to allow one party to opt-out, those judges and lawyers dependent on the lawsuits for work would be out of business. I just can't see the judges voluntarily writing themselves pink slips.
Personally, I think there's way too much litigation in the US, but this isn't the right way to reduce it. Especially in Sony's case, when fewer rootkits and more security admins might have been cheaper and have produced better results than hiring lawyers.
Look, like it or not, countries spy on other countries. Almost certainly the Russian stealth fighter and the Chinese attempts to build one are based on technical data either not erased off sold hard drives or "acquired" by various means. Unless the US knows specifically what was obtained, specifically how far these two nations have been able to process that data, and specifically what conclusions and resulting anti-stealth technology they've sold on, the US has no means to know if it even has a trump card.
That, almost certainly, is the reason they're not deployed. America doesn't know if the tech is still effective, nobody else knows if the countermeasures are still effective.
DeHavilland had chief executives as part of the test pilot crew. They still had fatal accidents, but strangely not as many as other companies. I think if military aircraft companies like Boeing and Lockheed Martin adopted a similar strategy, you'd see designs improving very rapidly indeed.
No software is ever going to be 100% bug-free, and there are diminishing returns on investing time and effort into testing and bug hunting. However, it is always prudent to ask if enough time and effort is being spent on such things. Are they developing the tests first and then writing code to the tests? Are the specifications up to par? Does the code have sane handling of errors (including non-fatal bugs)?
Doesn't surprise me in the least. I don't know what g the DH98 Mosquito could pull, but again I wouldn't be the least bit surprised if it was capable of taking out a pilot if mishandled. The Stuka, for all its limitations, was a good plane - the Germans had some exceptional engineering talent at the time in just about every area. It's a damn shame it wasn't put to better use - society could easily be decades further ahead than it is if those same brains had been put to less lethal pursuits.
Slashdot Mirror
I'd have to agree with you on optimizing. Chances are, though, that you could slim down further since not every page includes every feature of every library. The AJAX, CSS and other "enhancing" content never actually used on a given page never needs to be sent over with that page. "Page cleaning" before delivery should really be standard, IMHO -- what's the point of broadband if the page loads no faster than it did on a 56K modem?
(I've also seen far too many servers with page compression disabled because either the admin didn't know it could be enabled or because of bugs in the server.)
Of course, if Java hadn't had such limited communication with the browser page, the background libraries could be sent once for a site rather than once for a page, then cached locally on the client.
Precisely, and trying to shape human nature through regulation is where you run into the politicized nature of these environments. Yes, these systems should never be connected to the Internet (even by indirect means such as sneaker net) and yes fixing the problem must take human nature into account, but it must do so cleverly because applying the fix also involves human nature. That's a tough one.
To your example of someone stealing a car, I'd point out that yes the person who *stole* it may be guilty, but mistaken identity happens all the time and that means that you cannot go by one person's word alone that the person labelled as the thief is indeed the thief. There are all kinds of cases where vigilantes have attacked the wrong person for this very reason. Those vigilantes aren't part of the judicial system, so should they presume innocence? Yes. Obviously. Why obviously? Because the scientific method only works if you are trying to falsify things, it doesn't work if you are trying to prove things. (See Sherlock Holmes as to why this is important in criminal matters - he may be fictional, but he's a better role-model than Joe Thug.) If you start by assuming the person you think is guilty is, in fact, innocent then you can apply rigorous methods to falsify that hypothesis. Rigour is what matters.
To your example of AZ law, the Feds are investigating the sheriff in question (and seem on the verge of numerous indictments) because said sheriff DID regard that as sufficient grounds.
No, the police report only says what the police report says. Anyone can type out a document. Doesn't make the document true. Those of us from Britain are familiar with the Guildford 6 and the Birmingham 4, cases where false reports were made and falsified confessions (complete with forged signatures) were provided. We know that if these things could happen, they can happen and eventually will happen. You MUST therefore not accept any police report as being Absolute Truth, you must only accept it as being evidence offered that is no different from any other evidence, given no more or less weight, and scrutinized accordingly.
My own belief is that NO law, Federal or State, should be ignored, that ALL should be equal before the law, and that ALL evidence should be weighed and analyzed to the greatest practical* extent. Adversarial systems are something we're probably stuck with, and they serve a purpose, but monocultures are always defective and the current adversarial system in the US is definitely in the defective category.
*It would be nice if DNA evidence was analyzed by doing genomic decodes. Nice, but not practical. Maybe in 50 years. Perhaps. There will always be greater scrutiny technically possible, but you've got to draw some line or it gets absurd. In the case of DNA evidence, comparison of 8 markers isn't even remotely good enough - most labs wouldn't even break into a sweat if you upped the minimum to somewhere around 15-20. It would eliminate the false positives we're now seeing at no significant overhead in cost or time. That's an example of a practical threshold.
I seem to remember seeing SCADA vulnerabilities being added to vulnerability testing tools and IDS systems recently -- anyone know if this is related (ie: the tools now check for these non-existent flaws) or if the additions were to cover previously-reported bugs?
If the former, Siemens had best fix this damn fast. Infrastructure companies are in a corner - they don't have the cash for a major migration and alternative vendors are hardly thick on the ground. Some will be unable to afford decent security and others will be too politicized to secure their networks. Much of the infrastructure is too big and/or too expensive to duplicate, so the market is useless. The only place this can be fixed is at Siemens itself. The others that technically could won't and the rest can't.
The problem with the current paranoia over security is that you can't fix a fault you won't admit exists, companies won't deploy a fix if you tell them it's not needed, and so what you're ultimately left with is not security, merely obscurity.
Common sense is remarkably uncommon, but I guess you knew that as well. Pi is round, not square. And if I thought I was a guru, I'd be charging a hell of a lot more for my time. Spout off BS and I'll call you on your BS. If you don't want me calling you on it, don't spout it. It's very easy. You might even be able to achieve something like that.
Aether is the fifth element, so presumably only those who understand French sci-fi will understand the book.
That would explain why they're dead set against hackers rather than crackers. Gotta protect the Slytherins!
Of course, since there are now thousands of tunnels, the border patrol is also going to need drone C-130 transport planes kitted with synthetic-aperture ground-penetrating radar. Which can be done, but is probably going to add a bit to the cost.
Easy:
"Why should kids have all the fun? R/C toy for grown-ups includes all the accessories now found on the latest Real Action toys for children, in adult proportions. Real-action missiles come with realistic explosions."
Perhaps, perhaps not.
1. There may well be people who have US citizenship (and therefore be entirely legal) who have no documentation to prove it (and thus be undocumented). I'm not going to pretend that this would be a large number, but if there is even one such person then they are an undocumented immigrant not an illegal immigrant. This would include US legitimate tourists who have been robbed, people born out of the country with at least one legitimate US parent, etc. The former can apply to the consulate, but that assumes they're rational. Rational people are an endangered species. The latter may or may not have access to the consulate, even if theory says they should have.
2. In the US (not sure about your country), a person is presumed innocent until proven guilty in a court of law. Well, that's the theory, at least. Legal doctrine therefore states these people -cannot- be "illegal immigrants" until proven such. In part, this is due to (1), but it's also because you don't want some mad Arizonian sheriff arresting anyone who "looks funny" and deporting them without lawful right to do so. And, no, saying someone "looks funny" is not a lawful right.
3. Once a person is at a detention centre, there is NO evidence of where they were arrested. There are corrupt police - hopefully not many, but it's definitely non-zero. It would not take much to take a lawful US citizen from within the US and make it appear like they're illegally there, especially if said citizen has no documentation on them (ie: they're undocumented). It's entirely plausible for police to eliminate homelessness by dumping the homeless over the border, and for hospitals to eliminate mental illness the same way. (You've seen the stories on hospitals dumping patients in skid row.) These would not be illegals, these would be undocumenteds.
I cannot tell from a police report or a media photo whether the person was legally entitled to be in the US. Nor can you. Nor can anyone. That is why we have courts. Judge without knowledge at your peril, because it is inevitable that when a society converts a potential for a crime into a crime in itself, you WILL be judged without knowledge yourself. And that's a path that goes downhill FAST.
(Most of those who are passionate about convicting without prosecuting would do well to remember that the road to hell is paved with "good intentions" -- not intentions that are actually good, merely intentions you can fool yourselves into believing are good.)
CC would be good because it's designed for multiple types of data.
I just checked OpenCores (the first of the hardware open source sites) and the preferred license is LGPL on there. I would say that if it's already in use for hardware specs and is working for that then it's a viable license to use.
Ooooh! I like! Shouldn't be that hard - you have to have some sort of business license, right?
This has been discussed before on Slashdot and we've even had people from the US Patent Office discuss what it is the patent office actually does. I am not to blame if you have the memory of a goldfish, that is your problem and you are welcome to it. No, I won't post links, there's a search tool and there's google. Use 'em. I'm tired of you youngsters wanting us old folk to do the work you're too lazy to do. Now gerroff my lawn!
The USPO doesn't review patents. It makes them available in case anyone wants to complain (there's a year window for that) but they leave it up to the courts to handle the legality of patents. The problem is, the courts defer to the USPO and assumes they've handled the legality of patents. If someone actually took responsibility for patents, there probably wouldn't be the current mess.
...multi-tasking hardware has been around for a while. Hell, I started off on one such device (the PET 3032 with 4040 disk drive was multi-tasking as the 4040 had an independent CPU, ROM and RAM, so you could feed it an instruction and it would independently go away to do stuff without tying up the main CPU). I see nothing particularly new or innovative in the concept.
I dunno if the Amulet CPU can multi-task at the hardware level (versus OS multi-tasking in software) but the modular nature of it means I would not be at all surprised.
ftp://ftp.cs.man.ac.uk/pub/amulet/papers/jdg_asyn00.pdf
I still think that plaintiffs should be able to feed CEOs of malignant companies to the salt water crocodiles. The TV coverage would be so much better than most of the regular programming and it shouldn't require that many before corporate practices clean right up.
See "Refund Day".
That is something the courts have been hesitant to rule on but I believe that the US courts have ruled things like First Sale Doctorine, etc, take precedent over any such agreement. (ie: the EULA can't violate the law or otherwise impose a system contrary to the law.) Not being able to sue Sony in the event of Sony violating civil law - ie: denial of access to any system that can give relief - is usurping the courts entirely. This might not go down too well with judges, since if it's allowed, any product could have such a provision. If they allow one company to exempt itself from the legal system, they create a precedent (something judges are VERY loath to do) and case-law which would essentially state that any company could stipulate that a purchaser can't sue.
The civil court system depends heavily on people being able to sue each other. If the civil court system were to allow one party to opt-out, those judges and lawyers dependent on the lawsuits for work would be out of business. I just can't see the judges voluntarily writing themselves pink slips.
Personally, I think there's way too much litigation in the US, but this isn't the right way to reduce it. Especially in Sony's case, when fewer rootkits and more security admins might have been cheaper and have produced better results than hiring lawyers.
Look, like it or not, countries spy on other countries. Almost certainly the Russian stealth fighter and the Chinese attempts to build one are based on technical data either not erased off sold hard drives or "acquired" by various means. Unless the US knows specifically what was obtained, specifically how far these two nations have been able to process that data, and specifically what conclusions and resulting anti-stealth technology they've sold on, the US has no means to know if it even has a trump card.
That, almost certainly, is the reason they're not deployed. America doesn't know if the tech is still effective, nobody else knows if the countermeasures are still effective.
Apparently, it depends on which pilots you ask.
DeHavilland had chief executives as part of the test pilot crew. They still had fatal accidents, but strangely not as many as other companies. I think if military aircraft companies like Boeing and Lockheed Martin adopted a similar strategy, you'd see designs improving very rapidly indeed.
The thing that interests me is not the specifics of how the pilot lost consciousness but rather the computer element.
We know from the Quantas Airbus nosedive incident, the Airbus test flight crash, the Boeing computer crash resulting in an actual crash and similar that control software in aircraft is not at the standard one would hope. The Arianne 5 incident shows that corner-cutting (in that case, reusing software without testing it in the new context) is a hazardous approach in mission-critical systems.
No software is ever going to be 100% bug-free, and there are diminishing returns on investing time and effort into testing and bug hunting. However, it is always prudent to ask if enough time and effort is being spent on such things. Are they developing the tests first and then writing code to the tests? Are the specifications up to par? Does the code have sane handling of errors (including non-fatal bugs)?
Doesn't surprise me in the least. I don't know what g the DH98 Mosquito could pull, but again I wouldn't be the least bit surprised if it was capable of taking out a pilot if mishandled. The Stuka, for all its limitations, was a good plane - the Germans had some exceptional engineering talent at the time in just about every area. It's a damn shame it wasn't put to better use - society could easily be decades further ahead than it is if those same brains had been put to less lethal pursuits.
The problem with AI is that it can defect to the other side.
The cat ate the missiles midflight, so we shall never know if they were fitted with anti-gravity. The cat has been seen hovering mid-air, though.