[I'm pretty sure] that no scientist has ever proposed singularities as the source of ball lightning.
I'm pretty sure that Dr. Pace VanDevender, Vice President Emeritus of Sandia National Laboratories, with a Ph.D. in physics from the Imperial College of Science and Technology, University of London, counts as a scientist. He also has proposed singularities as the source of ball lightning, although he's not exactly confident (source requires subscription):
VanDevender himself knows it's on the wild side. "This is a long string of what-ifs," he admits, "it's very loosey-goosey." He reckons the odds of mini black holes existing are 1 in 10, the odds of catching one, maybe 1 in 1000, if he's optimistic. And that's what he wants to do: catch a black hole. "After stewing on it for years, I decided I did not want to die without knowing whether it was or was not real," he says.
it was all over/. a while ago... a team were able to create their own document and make its md5 match. JFGI.
I am perfectly aware that if you create your own document, you can then make an MD5 collision. I referred to that in each of my two previous posts:
There's also at least one known vulnerability in MD5, but it only allows you to construct two partly random strings that have the same hash, not to do any sort of preimage, so it's not relevant.
I'm not aware of any kind of systematic way to find a collision with an existing document aside from brute force. [note emphasis] [upgraded em->strong because blockquote doesn't like italics on/.?]
The files in question are not, however, your own document. They're a third-party document; you didn't create it. That means the exploit isn't relevant. It is not a preimage attack.
I suppose that theoretically the MPAA could have its members create the two different sets of files when creating the films, but that wouldn't really help either, unless Bittorrent prescribes a particular kind of file chunking. Each chunk is MD5ed, not the file as a whole, so if you change the size of one chunk by one byte you would throw off their entire game plan.
Furthermore, you can't create arbitrary collisions, as I recall: they have to share a prefix, and anything after that will be gibberish. That means the prefix will be the actual content of the movie, and so anyone who downloads either copy of the chunk will get the full content of the movie, just with a different string of gibberish tacked on. It might be possible to leverage this somehow, in a proprietary format, to stop one copy from running on standard media players, but that would require some kind of massive central directory, and in most cases you could just hack the software to drop the check.
How does the prosecution know that the perp thought his act was illegal?
He doesn't, and he doesn't have to. As far as I can tell, the MPAA just has to file suit for copyright infringement under the assumption that he committed some acts that were actually illegal, and as soon as they file suit they can generally get permission to demand information from you (such as the contents of your hard drive) relevant to the trial. FRCP rule 26(1) says that "For good cause, the court may order discovery of any matter relevant to the subject matter involved in the action."
I would note, though, that it's incredibly unlikely he'd be able to convince anyone that he actually thought he wasn't downloading a copyrighted work. Even "proof beyond a reasonable doubt" doesn't mean "proof so flawless that even the most hardened partisan couldn't find a way to explain it away".
but there have recently been studies and reports indicating an algorithmic short-cut to modifying data and yet leaving the md5 intact?
I haven't heard of those, but could be I missed them. I'm not aware of any kind of systematic way to find a collision with an existing document aside from brute force. Such a vulnerability, if practical to exploit, would make MD5 completely useless for any kind of cryptographic security, as far as I can tell, rather than merely for some kinds.
I suppose it's possible to generate random data until you can make it match the md5 hash or whatever torrent uses for block integrity
It's not, which is the entire point of MD5. There are 2128 possible combinations. Say you can use a supercomputer or whatnot to test, say, 250 (a bit over a quadrillion, a thousand million million) strings per second. This is ridiculously far past any reasonable capacity. Nevertheless, it will take you on average around 277 seconds to find a match. That's about 4,790,000,000,000,000 years, or several hundred thousand times the believed age of the universe.
Now, it's possible to reverse some MD5s using rainbow tables. That, however, depends on the string not being random gibberish, and it depends on getting the actual preimage rather than something that happens to have the same hash (which would defeat the purpose here). There's also at least one known vulnerability in MD5, but it only allows you to construct two partly random strings that have the same hash, not to do any sort of preimage, so it's not relevant.
if you put up a file for sharing named "star wars:episode 1" and it's just a blank file, aren't you comitting wire fraud?
I'm pretty sure you have to be asking for money or things of value for something to be fraud. If I try to sell you such a file, yeah, that's fraud. I'm pretty sure it's not if I just offer to give it away to you.
In this case, however, the so-called "artists" put up their copyrighted "works" (actually, just garbage, but as they created it, they actually do own the copyright to it) on a torrent server by themselves, free for the taking. They cannot then turn around and whine "you're stealing from us" when people do use the free service that they set up.
It's akin to a shop setting up a bin somewhere labeled "free samples", and then siccing the cops on those unsuspecting customers who "steal" from that bin...
Except for the fact that the downloader has no reason to believe that the person who put up the work actually has any rights to it. It's more akin to a sting operation, in concept at least: get them to do something they think is illegal, then prosecute them for it because it's evidence that they must have done similar things that are actually illegal. I don't know how the details will differ, though, in criminal versus civil.
I wonder tho... can you get burnt for downloading off those MPAA torrents ? Ive noticed that most of them have a large amount of peers/seeders which do not "connect" to you they only show up so lets say my torrent has 9000/3984 (9000 seeders 3984 peers) but none of them send you anything and if they do it's only parts of the file. Because it's not a working file can you get in trouble ? is the data even there ? or is it just fake muble jumble ? Does this mean I could get sued for labeling my video of my kitty sleeping "Rocky IV" and get sued ? How does this all work ? Unless you have a complete working copy of somthing they really can't do anything ?
Their primary tactic will be to threaten to bring you to court. If you actually get a lawyer, they might drop you for easier prey, of course. Their legal strategy would likely be to hold up your download as, while not necessarily against the law, evidence of a tendency to break the law, and therefore grounds for discovery proceedings that would allow them to rummage through your hard drive and find all the actual copyrighted stuff you downloaded. I suppose, anyway; IANAL.
This is sounding more and more like a Mafia story than anything else (and the mafiaa joke is still funny).
With the slight difference that rather than threatening illegal physical harm to you, the MPAA is threatening to file suit against you in a court of law for clear-cut infringement of legally-binding acts passed by Congress. I don't know about you, but even if I disagree with a law, I don't view it as immoral to enforce it.
How have the courts/law enforcement allowed this to proceed? Isn't this harassment?
Of course not. It's no more harassment to threaten suit given legal cause (and there is clear legal cause here) than it is for the police to chase down drug dealers or whatever. The MPAA is upholding the law, not breaking it, whether you like the law or not.
What if I legally own the movie and just want to download an XviD version for my mp3 player, cause I suck at ripping DVDs? How is that illegal?
It is illegal. You're reproducing the copyrighted work, which is illegal under 17 USC 106(1). Personal use is not fair use, in general. Ripping DVDs is also typically illegal. You are permitted to make copies for "archival purposes only" (17 USC 117(a)(2)), not for use in other media.
AFAIK, however, the MPAA makes a point of not going after legal owners of the media. If they found that you were a legal owner, they would probably drop you for a target whose prosecution would be better PR.
I suspect that they would, unless the defendant had some clever scenario that accounted for how the MPAA got evidence that they downloaded. People probably aren't going to believe that the MPAA will outright fabricate evidence, and it probably won't, not when it's so easy to get real evidence and not when the penalties for being caught would be so huge.
In "computer terms", you're copying the data, but you're not making a copy in the traditional sense. I'm not saying you're wrong about it still being a civil offense, but your logic is incorrect.
You are causing a copy of the data to appear on your hard drive. That you are using a particular machine or process to do so is irrelevant. The provider of the torrent is complicit, of course, for encouraging you to make the copy, but you're still the one causing that particular copy to be made. That is very much an infringement of copyright.
How much overhead is required by law? I should imagine any decent-sized charity would have to pay something for accountants, lawyers, and operating facilities. And from a cost-effectiveness perspective, surely good advertising is more valuable than actually giving the ad money to the poor people, because it nets you more money spending half of your income on ads is good if those ads pay for themselves at least twice over. And so on.
Overhead is not and should not be the major gauge of how charitable a charity is.
When people say "bad code", in my experience, they usually mean hard to read or understand. It's obviously not too quantifiable as a concept. The code should follow usual conventions on whitespace, it should use descriptive variable names, it should avoid overuse of complex syntax (ternary operators, nesting, recursion), it should avoid unnecessary or redundant steps, it should comment exactly those things that might be puzzling, functions should be broken up if they get too large. Basically, another programmer should be able, upon reading it, to know what it does and how to change it to do what he wants with a minimum of confusion and effort.
It's something that comes with experience, mainly. You'll hopefully figure it out once you've had to deal with enough bad code yourself.
Also, the compiler may know more CPUs than you do. For example, do you know the pairing rules for instructions on an original Pentium? The differences one must pay attention to when optimizing for an Thoroughbred Athlon vs. a Prescott P4 vs. a Yonah Pentium-M vs. a VIA Nehemiah? GCC does a pretty good job of generating optimized assembly code for each of these from the same C source code. If you were to do the same in assembly, you would have to write separate code for each CPU, and know the subtle differences as well as the compiler does.
I must admit to only having taken a semester's worth of assembly, but it seems to me that the major optimization is simply condensing verbose instructions. If you're trying to play fancy tricks with pipelines, sure, you'll have to do it differently, but that's a losing game.
How does gcc know what chip the program is going to be run on, anyway? Once it's compiled, it can be run on any processor that uses the given instruction set.
Well every advance in technology so far has led to an eventual advance in ethics and morals.
Well, not exactly. Or rather, we today would view it as an advance in ethics and morals, but our ancestors would have viewed it as a degradation of ethics and morals. Who's to say that aliens would be more like us than our ancestors? Is there really some major trend in our morality that can be taken to be associated with technology?
Sure, there's democracy . . . but Rome certainly became a lot less democratic with the ascent of Caesar to the throne, even through centuries more of technological advances; China is rapidly advancing technologically, but is getting barely less oppressive; and so forth. If anything, I would say that democracy leads to technology, due to the spirit of freedom to explore the uncharted, not the other way around. The principles of the Enlightenment led to both liberty and scientific progress, the latter didn't lead to the former.
What other trends have there been in morality? Less harm to one's enemies? Okay, sure, I can see that. More technology allows gentler treatment by those in power without jeopardizing that power. But you have to have the desire to treat your enemies nicely first. I used the example of ants as creatures with no empathy, who place the importance of the group far above that of any number of individuals. It would be quite possible for an ant-like species to gain enough intelligence to reach us. As long as you have empathy for group members, you can prosper even while having no concern for non-group members (say, us).
The fact that we are acting like animals though, with regard to torture and our current malaise towards actual democracy, means that very likely the aliens view us more as such, than say if we still were following the constitution, for one small example.
Again, you're making assumptions about how aliens would perceive things that I think are unwarranted. Ants follow their queen, and have no compunction about killing those few members of the colony who don't. They certainly aren't big on democracy. That doesn't mean that ant-like drones (who cannot reproduce) need be mindless or unintelligent, just have different motives: they would have to place the good of the colony over anything else. Thus they could still invent advanced technology, it would just be to advance the hive and their curiosity rather than themselves and their curiosity. Primates, essentially all of whom can reproduce, often favor themselves or their friends or strangers rather than their groups.
As another example, consider computers. Computers are, in certain respects, more intelligent than any human, and in all other respects they're increasingly approaching the intelligence of humans, but they do nothing but what they're told. This perhaps above all shows the separation between capability and motive. Intelligence and curiosity need not come with even the slightest hint of compassion, and technology requires little more than intelligence, curiosity, and some degree of cooperation with others. It's perfectly possible, if not in fact likely, that we and any theoretical aliens would each view the other as depraved.
Re:You have no idea what you're talking about
on
SQL Hacks
·
· Score: 1
just like a good C compiler normally produces better and faster assembly language than a human can normally produce.
That is flat-out incorrect. Any seriously performance-intensive programming such as computer games, C library functions, drivers, and some parts of operating systems is still done in assembly (or compiled in C and then optimized in assembly) because it's a lot faster and a lot more compact. Why would Microsoft bother putting an inline assembly feature in VC if it were slower than what the compiler produced?
If you want a more specific figure, the ballpark figure I was given at the beginning of my assembly course is that assembly is around three times faster to execute and ten times slower to write than equivalent C code. A good C compiler is hardly "better and faster" than manually-written assembler.
Luckily, any decent compiler will warn you on that (and those who ignore warnings should be summarily executed, they're there for a reason).
Unfortunately, by that metric PHP does not have a good compiler. Python, of course, won't allow assignment in if conditions to begin with, which I think goes rather too far. Possibly my favorite solution is to use some special syntax for assignment, like:= instead of just =. But the idea of getting in the habit of using 0 == x instead of x == 0 is a good one, definitely, although it doesn't work for comparison of two variables. (Actually, I believe 0 == $x is even faster than $x == 0 in PHP, although I could be wrong.)
Consider that the drive I bought at Costco 10 years ago(500MB) costs on the order of 500 to 1000 times more (that'salmost two magnitudes) than storagetoday, and that Microsoft continues to charge at the same rate --they even seem to adjust for inflation.
Hardware develops at exponential rates, software develops incrementally. You're still paying effectively the same price for a product that, by and large, has more features (even if it's more gluttonous in terms of resource usage). Expecting a product improved by a factor of a thousand is unrealistic in almost any industry. Do you expect your car to drive a thousand times more safely and quickly than it did ten years ago?
I don't dispute that Microsoft puts out a fairly shoddy and overpriced product. It's a monopoly, after all. But expecting any software to improve as quickly as hardware is extremely unrealistic.
You're using a definition of "advanced" rather at odds with what everyone else here means by "advanced". We're all talking about technologically advanced, whereas you seem to be talking about morally advanced. But that seems like an odd thing to expect of aliens in any case: surely their morals would be dramatically different from ours?
It seems strange to expect that very many aliens would disapprove of torture, for instance. Why would they? We haven't, up until perhaps fifty to a hundred years ago. You can make a convincing Machiavellian argument that the ends justify the means and a few probable (even if not certain) terrorists' suffering is justified. That kind of argument is mostly out of favor now, but I very much doubt it would be if we were, say, ants.
Unless, of course, it's heavily compressed. Compression, after all, increases apparent randomness. Order implies redundancy if I can guess what the next bit of information is going to be, it's a waste of bandwidth to send a whole bit to tell me. Send part of a bit if my guess is right, more than a bit if I'm wrong, and you'll save on bandwidth.
For instance, in the English language Q is almost always followed by U. This is nonrandom. As soon as I receive a Q from you, I can be almost certain that I'll get a U next. So instead of wasting that extra letter, we can save a bit of bandwidth by deciding that "Q" means "QU" and "QU" means "Q". Then Q is no longer usually followed by U, which increases the randomness and reduces the size.
We might still be able to figure it out, but I don't know if it's a sure thing. It could be I can't claim to be an expert on compression.
As a C/C++ developer I am a little offended by the article summary. Certainly C/C++ has a lot of flexibilities that allow bad developers to write bad code. However, many other languages, e.g. Java, allow bad programmers to write code that looks good because of stronger type checking, reduced use of pointers and the like. However, nothing stops a bad developer from writing insecure code in any language. Maybe you don't manage your resources correctly. Maybe you do a bad job of implementing encryption/protected storage. Maybe your authentication scheme is weak, your site is vulnerable to cross-site scripting vulnerabilities, or your session data can be easily spoofed.
Yes . . . but all those are true in C/C++, too, in addition to things like buffer overflows. There are fewer opportunities for serious error, and assuming you stay away from eval()-like functions, in interpreted languages there are no opportunities whatsoever for arbitrary code execution no matter how hard you try. If you're an expert and careful C/C++ programmer, sure, you'll write much more secure code than a novice and hasty Java programmer. But people of comparable expertise and carefulness will write more secure code in languages other than C/C++, and they'll probably write it faster too. The only major disadvantage is execution speed.
Slashdot Mirror
I'm pretty sure that Dr. Pace VanDevender, Vice President Emeritus of Sandia National Laboratories, with a Ph.D. in physics from the Imperial College of Science and Technology, University of London, counts as a scientist. He also has proposed singularities as the source of ball lightning, although he's not exactly confident (source requires subscription):
I am perfectly aware that if you create your own document, you can then make an MD5 collision. I referred to that in each of my two previous posts:
The files in question are not, however, your own document. They're a third-party document; you didn't create it. That means the exploit isn't relevant. It is not a preimage attack.
I suppose that theoretically the MPAA could have its members create the two different sets of files when creating the films, but that wouldn't really help either, unless Bittorrent prescribes a particular kind of file chunking. Each chunk is MD5ed, not the file as a whole, so if you change the size of one chunk by one byte you would throw off their entire game plan.
Furthermore, you can't create arbitrary collisions, as I recall: they have to share a prefix, and anything after that will be gibberish. That means the prefix will be the actual content of the movie, and so anyone who downloads either copy of the chunk will get the full content of the movie, just with a different string of gibberish tacked on. It might be possible to leverage this somehow, in a proprietary format, to stop one copy from running on standard media players, but that would require some kind of massive central directory, and in most cases you could just hack the software to drop the check.
He doesn't, and he doesn't have to. As far as I can tell, the MPAA just has to file suit for copyright infringement under the assumption that he committed some acts that were actually illegal, and as soon as they file suit they can generally get permission to demand information from you (such as the contents of your hard drive) relevant to the trial. FRCP rule 26(1) says that "For good cause, the court may order discovery of any matter relevant to the subject matter involved in the action."
I would note, though, that it's incredibly unlikely he'd be able to convince anyone that he actually thought he wasn't downloading a copyrighted work. Even "proof beyond a reasonable doubt" doesn't mean "proof so flawless that even the most hardened partisan couldn't find a way to explain it away".
I haven't heard of those, but could be I missed them. I'm not aware of any kind of systematic way to find a collision with an existing document aside from brute force. Such a vulnerability, if practical to exploit, would make MD5 completely useless for any kind of cryptographic security, as far as I can tell, rather than merely for some kinds.
It's not, which is the entire point of MD5. There are 2128 possible combinations. Say you can use a supercomputer or whatnot to test, say, 250 (a bit over a quadrillion, a thousand million million) strings per second. This is ridiculously far past any reasonable capacity. Nevertheless, it will take you on average around 277 seconds to find a match. That's about 4,790,000,000,000,000 years, or several hundred thousand times the believed age of the universe.
Now, it's possible to reverse some MD5s using rainbow tables. That, however, depends on the string not being random gibberish, and it depends on getting the actual preimage rather than something that happens to have the same hash (which would defeat the purpose here). There's also at least one known vulnerability in MD5, but it only allows you to construct two partly random strings that have the same hash, not to do any sort of preimage, so it's not relevant.
That's why their lawyers would weed you out of the jury pool.
I'm pretty sure you have to be asking for money or things of value for something to be fraud. If I try to sell you such a file, yeah, that's fraud. I'm pretty sure it's not if I just offer to give it away to you.
Copyright infringement isn't even a crime except for under particular cases.
No, it's a cause of action for civil suit. That's why the MPAA has to enforce it, and can't really rely on the police.
It's akin to a shop setting up a bin somewhere labeled "free samples", and then siccing the cops on those unsuspecting customers who "steal" from that bin...
Except for the fact that the downloader has no reason to believe that the person who put up the work actually has any rights to it. It's more akin to a sting operation, in concept at least: get them to do something they think is illegal, then prosecute them for it because it's evidence that they must have done similar things that are actually illegal. I don't know how the details will differ, though, in criminal versus civil.
Ive noticed that most of them have a large amount of peers/seeders which do not "connect" to you they only show up so lets say my torrent has 9000/3984 (9000 seeders 3984 peers) but none of them send you anything and if they do it's only parts of the file. Because it's not a working file can you get in trouble ? is the data even there ? or is it just fake muble jumble ? Does this mean I could get sued for labeling my video of my kitty sleeping "Rocky IV" and get sued ? How does this all work ? Unless you have a complete working copy of somthing they really can't do anything ?
Their primary tactic will be to threaten to bring you to court. If you actually get a lawyer, they might drop you for easier prey, of course. Their legal strategy would likely be to hold up your download as, while not necessarily against the law, evidence of a tendency to break the law, and therefore grounds for discovery proceedings that would allow them to rummage through your hard drive and find all the actual copyrighted stuff you downloaded. I suppose, anyway; IANAL.
With the slight difference that rather than threatening illegal physical harm to you, the MPAA is threatening to file suit against you in a court of law for clear-cut infringement of legally-binding acts passed by Congress. I don't know about you, but even if I disagree with a law, I don't view it as immoral to enforce it.
How have the courts/law enforcement allowed this to proceed? Isn't this harassment?Of course not. It's no more harassment to threaten suit given legal cause (and there is clear legal cause here) than it is for the police to chase down drug dealers or whatever. The MPAA is upholding the law, not breaking it, whether you like the law or not.
It is illegal. You're reproducing the copyrighted work, which is illegal under 17 USC 106(1). Personal use is not fair use, in general. Ripping DVDs is also typically illegal. You are permitted to make copies for "archival purposes only" (17 USC 117(a)(2)), not for use in other media.
AFAIK, however, the MPAA makes a point of not going after legal owners of the media. If they found that you were a legal owner, they would probably drop you for a target whose prosecution would be better PR.
How is that hearsay? Hearsay is "Evidence based on the reports of others rather than the personal knowledge of a witness", but this testimony is based on the personal knowledge of the witness (the MPAA). It's up to the jury, and possibly to some extent the judge, to decide whether the witness is credible.
I suspect that they would, unless the defendant had some clever scenario that accounted for how the MPAA got evidence that they downloaded. People probably aren't going to believe that the MPAA will outright fabricate evidence, and it probably won't, not when it's so easy to get real evidence and not when the penalties for being caught would be so huge.
You are causing a copy of the data to appear on your hard drive. That you are using a particular machine or process to do so is irrelevant. The provider of the torrent is complicit, of course, for encouraging you to make the copy, but you're still the one causing that particular copy to be made. That is very much an infringement of copyright.
How much overhead is required by law? I should imagine any decent-sized charity would have to pay something for accountants, lawyers, and operating facilities. And from a cost-effectiveness perspective, surely good advertising is more valuable than actually giving the ad money to the poor people, because it nets you more money spending half of your income on ads is good if those ads pay for themselves at least twice over. And so on.
Overhead is not and should not be the major gauge of how charitable a charity is.
When people say "bad code", in my experience, they usually mean hard to read or understand. It's obviously not too quantifiable as a concept. The code should follow usual conventions on whitespace, it should use descriptive variable names, it should avoid overuse of complex syntax (ternary operators, nesting, recursion), it should avoid unnecessary or redundant steps, it should comment exactly those things that might be puzzling, functions should be broken up if they get too large. Basically, another programmer should be able, upon reading it, to know what it does and how to change it to do what he wants with a minimum of confusion and effort.
It's something that comes with experience, mainly. You'll hopefully figure it out once you've had to deal with enough bad code yourself.
I must admit to only having taken a semester's worth of assembly, but it seems to me that the major optimization is simply condensing verbose instructions. If you're trying to play fancy tricks with pipelines, sure, you'll have to do it differently, but that's a losing game.
How does gcc know what chip the program is going to be run on, anyway? Once it's compiled, it can be run on any processor that uses the given instruction set.
Well, not exactly. Or rather, we today would view it as an advance in ethics and morals, but our ancestors would have viewed it as a degradation of ethics and morals. Who's to say that aliens would be more like us than our ancestors? Is there really some major trend in our morality that can be taken to be associated with technology?
Sure, there's democracy . . . but Rome certainly became a lot less democratic with the ascent of Caesar to the throne, even through centuries more of technological advances; China is rapidly advancing technologically, but is getting barely less oppressive; and so forth. If anything, I would say that democracy leads to technology, due to the spirit of freedom to explore the uncharted, not the other way around. The principles of the Enlightenment led to both liberty and scientific progress, the latter didn't lead to the former.
What other trends have there been in morality? Less harm to one's enemies? Okay, sure, I can see that. More technology allows gentler treatment by those in power without jeopardizing that power. But you have to have the desire to treat your enemies nicely first. I used the example of ants as creatures with no empathy, who place the importance of the group far above that of any number of individuals. It would be quite possible for an ant-like species to gain enough intelligence to reach us. As long as you have empathy for group members, you can prosper even while having no concern for non-group members (say, us).
The fact that we are acting like animals though, with regard to torture and our current malaise towards actual democracy, means that very likely the aliens view us more as such, than say if we still were following the constitution, for one small example.Again, you're making assumptions about how aliens would perceive things that I think are unwarranted. Ants follow their queen, and have no compunction about killing those few members of the colony who don't. They certainly aren't big on democracy. That doesn't mean that ant-like drones (who cannot reproduce) need be mindless or unintelligent, just have different motives: they would have to place the good of the colony over anything else. Thus they could still invent advanced technology, it would just be to advance the hive and their curiosity rather than themselves and their curiosity. Primates, essentially all of whom can reproduce, often favor themselves or their friends or strangers rather than their groups.
As another example, consider computers. Computers are, in certain respects, more intelligent than any human, and in all other respects they're increasingly approaching the intelligence of humans, but they do nothing but what they're told. This perhaps above all shows the separation between capability and motive. Intelligence and curiosity need not come with even the slightest hint of compassion, and technology requires little more than intelligence, curiosity, and some degree of cooperation with others. It's perfectly possible, if not in fact likely, that we and any theoretical aliens would each view the other as depraved.
That is flat-out incorrect. Any seriously performance-intensive programming such as computer games, C library functions, drivers, and some parts of operating systems is still done in assembly (or compiled in C and then optimized in assembly) because it's a lot faster and a lot more compact. Why would Microsoft bother putting an inline assembly feature in VC if it were slower than what the compiler produced?
If you want a more specific figure, the ballpark figure I was given at the beginning of my assembly course is that assembly is around three times faster to execute and ten times slower to write than equivalent C code. A good C compiler is hardly "better and faster" than manually-written assembler.
Unfortunately, by that metric PHP does not have a good compiler. Python, of course, won't allow assignment in if conditions to begin with, which I think goes rather too far. Possibly my favorite solution is to use some special syntax for assignment, like := instead of just =. But the idea of getting in the habit of using 0 == x instead of x == 0 is a good one, definitely, although it doesn't work for comparison of two variables. (Actually, I believe 0 == $x is even faster than $x == 0 in PHP, although I could be wrong.)
Hardware develops at exponential rates, software develops incrementally. You're still paying effectively the same price for a product that, by and large, has more features (even if it's more gluttonous in terms of resource usage). Expecting a product improved by a factor of a thousand is unrealistic in almost any industry. Do you expect your car to drive a thousand times more safely and quickly than it did ten years ago?
I don't dispute that Microsoft puts out a fairly shoddy and overpriced product. It's a monopoly, after all. But expecting any software to improve as quickly as hardware is extremely unrealistic.
You're using a definition of "advanced" rather at odds with what everyone else here means by "advanced". We're all talking about technologically advanced, whereas you seem to be talking about morally advanced. But that seems like an odd thing to expect of aliens in any case: surely their morals would be dramatically different from ours?
It seems strange to expect that very many aliens would disapprove of torture, for instance. Why would they? We haven't, up until perhaps fifty to a hundred years ago. You can make a convincing Machiavellian argument that the ends justify the means and a few probable (even if not certain) terrorists' suffering is justified. That kind of argument is mostly out of favor now, but I very much doubt it would be if we were, say, ants.
Unless, of course, it's heavily compressed. Compression, after all, increases apparent randomness. Order implies redundancy if I can guess what the next bit of information is going to be, it's a waste of bandwidth to send a whole bit to tell me. Send part of a bit if my guess is right, more than a bit if I'm wrong, and you'll save on bandwidth.
For instance, in the English language Q is almost always followed by U. This is nonrandom. As soon as I receive a Q from you, I can be almost certain that I'll get a U next. So instead of wasting that extra letter, we can save a bit of bandwidth by deciding that "Q" means "QU" and "QU" means "Q". Then Q is no longer usually followed by U, which increases the randomness and reduces the size.
We might still be able to figure it out, but I don't know if it's a sure thing. It could be I can't claim to be an expert on compression.
Good point. Allofmp3.com seems to mostly get away with it, and that has an entire industry fighting it.
Yes . . . but all those are true in C/C++, too, in addition to things like buffer overflows. There are fewer opportunities for serious error, and assuming you stay away from eval()-like functions, in interpreted languages there are no opportunities whatsoever for arbitrary code execution no matter how hard you try. If you're an expert and careful C/C++ programmer, sure, you'll write much more secure code than a novice and hasty Java programmer. But people of comparable expertise and carefulness will write more secure code in languages other than C/C++, and they'll probably write it faster too. The only major disadvantage is execution speed.