The author of a document has certain rights that are described in copyright law. For example, the right to produce derivative works or the right to distribute copies. As Lessig notes in the interview, consumers can also enter into agreements with copyright holders via things like shrink-wrap licenses. Entering into a restrictive agreement with someone is not tantamount to surrendering the rights an author has as copyright holder. Although it might strike some chord of fairness, copyright holders don't lose these rights just because you agreed not to read their books out loud (or whatever), any more than they lose their rights when they agree to let a publisher produce and sell copies of their work.
Just because the GPL doesn't grant certain privileges, that's no reason you can't obtain them anyway. Ask the code's author.
The existence of the GPL makes it easy for people to license their code under a certain set of terms. Just because those terms don't include the terms you want, that doesn't make it infectious. It's only an infection if you're too lazy to ask the author of the code for the permissions you want (that aren't included in the GPL), or if authors who are too lazy to answer their email fail to give permission for uses they would have deemed acceptable. If the author of the code would never say no to such a request, then sure, they should have released their code under a more permissive license. If they ever say no, then the GPL is actually protecting something they consider important. If the author is unreachable, then you have to accept their choice of the GPL at face value -- those are exactly the permissions they intended to grant.
The GPL only seems to infect everything it touches. If you want to use GPLed code but can't be GPL-compatible, nothing prevents you from contacting the author of the code for permission. Well, at least for smaller projects with easily reachable authors. For code that has hundreds of authors, some deceased, this might be more difficult.
Indeed, this is huge. With this kernel, I believe NFS (esp NFSv3) moves from something you either have to pray your distribution vendor got right, or something you have to cobble together from patches, to a snazzy and high-functioning part of the Linux kernel right out of the box. Many thanks are owed to Trond Myklebust, David Higgen, Alan Cox, and the rest of the NFS gang.
I think the new Dune miniseries clearly owes more to the old Batman TV series than to Frank Herbert's or David Lynch's work. Although there have been a few parts that seemed well-dramatized, for the most part the zany costumes and comically exaggerated overacting (from both accomplished actors and unknowns, so I have to assume it was a decision of the director) scream camp. Although the novel has its flaws, and sure, in places it lends itself to a campy adaptation (esp. the earlier scenes with the Baron), the miniseries looks like it will pass on the opportunity to draw from the depth of the novel.
I suppose I am obligated to indicate how many times I've read Dune and the rest of the series. I have no idea though. Probably 30 for Dune, average of 5 for the rest.
dan
ps my subject line is not intended as a reference to the use of the word "batman" to refer to the character Patrin who shows up in heretics of dune.
The article about the Natapoff argument on the desirability of the electoral college system was interesting. It was a little light on details, but he seems to argue from the point of view that the only desideratum in designing an election system is maximizing individual power in voting. This seems like an assumption, not a research conclusion, and I wonder how widely it's shared. In light of the fact that Natapoff's quoted comments are, to say the least, deeply condescending to basically everyone, this seems like an important point. If he's going to claim to have proved the electoral college system is better than going by popular vote, at a minimum it seems like he ought to have a definition of "better" that is universally acceptable. Is the complete work available online anywhere?
Incidentally, it seems a little suspect to argue that democracy hinges on this system, when in the majority of elections it makes no difference whatsoever in the outcome. Correctly or not, people seem to feel they would have more voting power if votes for the second or third place candidate in your state could actually make a difference.
You are mistaken that patents do not spur research. If there were no patent protection, there would be no big drug companies. Drug companies sink an astronomical amount of money into obtaining and defending patents, and practically every day some drug company's stock shoots up or bottoms out because a patent is either obtained, upheld or shot down. We can argue about how good a job these drug companies do at developing drugs. We might even speculate about unethical machinations behind the "research" used to gain approval. But there's no question that patent protection underlies the entire industry. You don't sink billions of dollars into developing zoloft if it's going to sit between a generic equivalent and a better-marketed competitor's version on the shelf. And you can't keep a drug's composition a trade secret.
It's hardly different with copyright. For the time being, books are put out by publishers, and publishers stay in business because of copyright protection. A few publishers try to make money producing new editions of work that has run out of copyright protection. But it's pretty obvious from the distribution of books in bookstores that this is not where the money is. I think it's safe to say that Steven King's publishers would be a lot less thrilled about printing up millions of copies of his books if they knew that there would be competing editions within a week. And of course, say goodbye to hardcover editions.
I do have some personal interest in copyright protection, since I'm self-publishing a book. If there were no copyright protection, someone with more money than I have could easily produce a more attractive edition and sell it for less. But I still don't know if I really think copyright protection is a good thing. It runs against all the pure motivations for writing a book, and in my opinion there are too many people out there writing books for the wrong reasons.
In the case of patents, though, despite my reservations about the ethics, let alone the integrity, of the big drug companies, I think some patent protection is valuable. There are things I think we as people should want to have, that we can have a lot sooner if we compensate inventors with patent protection. Although it's true that coming up with a brilliant idea second is a real problem, this should rarely happen. Patents are supposed to apply to non-obvious inventions. Alas, the USPTO doesn't always see it that way. This doesn't mean that patent protection isn't valuable or desirable, it only means that it needs some tweaking.
I wanted to order the CmdrTaco version, with all those DVD's, MP3's, Simpsons and South Park episodes pre-loaded, but I couldn't find the ordering page on CmdrTaco.net. Please advise.
It's not normally my inclination to try to re-allocate other people's efforts. After all, as much as the world needs biomedical engineers, it would be a shame if there weren't just a few people who devoted their lives to figuring out where all my socks go after I put them in the washer.
That said, some projects just scream out, "I have lots of energy and resources but I couldn't figure out anything useful to do with them." The web, for example. And possibly pi digit identification. So wouldn't it be nice to start a project to help herd people with that kind of muscle into projects they and everyone else would find more useful? Far be it from me to suggest a particular project that should have been diverted (for all I know this pi thing involved new frontiers in computational methods). But I'm sure there are lots of projects out there, in many fields, that haven't gotten done because nobody had both the interest and the resources.
I'd consider hosting such a project's web pages myself, I'm sure they would be extremely low-volume.
Not being one to trust anyone on anything, a few years ago, I (along with a colleague) decided to embark on a little experiment to see just how easy it was to render a floppy unreadable. We were concerned about everyday risks (and this guy happened to be doing some research that involved small electric motors). So we tried leaving them on top of monitors for a few minutes, leaving them on top of speakers, and moving the magnet end of the motors across the surface of the disk at various angles. There was one small Word file on this disk (on a Mac filesystem), and we used the same disk throughout. Despite repeated trials, especially with the otherwise powerful (in the paperclip sense) motor magnets, we were completely unable to erase the file or damage the filesystem.
I believe at the time our scientific conclusion was that floppies are not based on magnetism, but on "tiny bubbles of ectoplasmic phlogiston." We never tried the condition where the file was the sole copy of some critical document, I don't know if that would have affected the results.
"Implementing Windows emulation on Linux" isn't part of the plex86 picture. Like VMWare it allows you to run multiple operating systems simultaneously on one machine. Or rather two machines, one real one virtual. You still need to have Windows installed, you just don't have to reboot to use it.
I can imagine a few reasons why software like Plex86 or VMware might ultimately harm Microsoft. For example, they might help ease a Windows-to-Linux transition. But they don't let you run Windows software without Windows. Even Wine doesn't really do that (yet).
The RSA algorithm has been public knowledge since it was developed. It's release (a few weeks in advance of the patent expiration) simply means people can use it without a license from RSA.
RSA has not been cracked. Some specific RSA keys of particular lengths (e.g., 512 bits) have been discovered. That's no big deal, since we already know roughly how much computational power it should take to crack a given key. And some weaknesses in particular implementations of RSA have been noted. But it's reasonably well understood how much (implementation-independent) security is provided by a given key length, and notwithstanding advances in factoring, that has stood up pretty well.
RSA may or may not have something better but top secret up their sleeves, but if so it hasn't been exposed to the scrutiny of the RSA algorithm. And the most likely areas for improvement are in computational efficiency and things like that, not in security per se.
"HaX0rz are out there and eventually they will get in one way or another...Is that hopeless?"
Well, if you don't mind a little freeform juxtaposition, then yes! The real question is whether or not it's excessively hopeless, and I think the answer is yes there too. Security isn't easy, but it's possible to restrict risk to acceptable levels. I agree that backups and insurance are also important. But anyone who doesn't have these already knows they're incompetent.
And the insurance really shouldn't be so expensive for anyone who is conscientious about risk. To put this a little differently, I wonder how an insurance policy would sell if it had two tiers of coverage, the lower of which kicked in if it were found that at the time of the incident a relevant patch more than five days old was available. Well-run organizations would be happy to buy it, poorly run ones would have to stick with the expensive stuff.
Bruce Schneier seems like a pretty conscientious guy in print. But this article just reads like a detailed ad four counterpane's services.
In connection with his new book (which I haven't read yet, because I'm still trying to find a good consultant to find me a morally upstanding bookseller), I wonder how much of his attitude is a necessary contingency of running a security business, or if that's why he started counterpane in the first place. I don't find fault with his presentation of facts, more with the sense of hopelessness he has conveyed in recent writing (I'm going mostly by articles, excerpts, and his crypto-gram newsletter).
It seems like the problem to be solved here is, in reduced form, telling the difference between pages that are actually more relevant to the search criteria than others and pages that are trying to fake being especially relevant. That sounds like an AI-complete problem (actually it sounds a whole lot like a web page Turing test). In that case obscurity might be the best model, since the best solutions will probably be exploitable. (Actually, even real intelligence is probably exploitable.)
Zappa's Law (as relayed by StoryMan, who admits he may have gotten it wrong) may be cute, but it will get you into some serious trouble if you actually believe it in some consequential context. Let me give two pertinent examples.
It's true that with respect to dying on a given flight, there are two major outcomes: dying and not. However, those two outcomes don't come in even proportions -- they aren't 50-50. If people from my country always fly on 50-50 Air, and people from your country fly on 99.99999-.00001 Air (both pretty shoddy organizations), then after a short period of time my country will be depopulated, while yours will thriving (but perhaps a bit nervous).
Now let's look at two online booksellers. 50-50 books figures either a book sells or it doesn't. PriceSetter Books is interested in the factors that contribute to the probability of a sale. PriceSetter figures out it can double it's sales of "Lucky Numbers" by changing the price from $13 to $12.95, and at the same time sell the same number of copies of "Worthless at Any Price" at $15 as it did at $12. 50-50 figures no matter what the price is, there are still exactly two outcomes, and just charges whatever they think is a good price. In both cases, a given consumer either buys the book or doesn't. But PriceSetter will earn more money, because they know that when you're talking about a million customers, there is a huge difference between 3% and 3.1%. And neither is all that close to 50-50.
Nobody really expects that booksellers won't try to maximize their profits by setting prices appropriately, whether profit is maximized in the short term, long term, or whatever (selling a book at a loss could maximize profits in some cases, I'm sure). But obviously in this case, people aren't happy with how Amazon went about it. I'm not so happy about it either, for many of the reasons listed in this thread.
Anyway, relevant to your message, I think we're stuck with the ad men and marketing slimeballs, because even if they're completely incompetent and come up with the wrong answers (price testing increases profits by.02%, but puts your company out of business when it gets posted on slashdot), it's nonetheless true that price shenanigans, heavy graphics, and all the stuff you and i may hate, have some connection with profits, and it's going to be hard to convince Amazon to stop trying to figure out what that connection is.
It seems like the only thing they could possibly be referring to as their intellectual property here is their specific (and earth-shatteringly trivial) encoding of the bar codes. There is certainly nothing innovative about the encoding technique. So it seems roughly equivalent to claiming a particular number as your intellectual property. I'd like to read the patent, but I'm already at capacity for patent-related outrage this week.
On the other hand, the bar code reader aside (which I've actually found helpful in testing some bar codes I was printing), I think the digital convergence stuff (meaning mainly the airwaves signal stuff) is kind of neat. It seems like a primitive but positive step towards some kind of actual (for lack of a better word) convergence. Not that I want any right now, but I still think it's cool.
(Also, is it true that you have to defend a patent or risk it falling into the public domain? I thought that was only true of trademarks.)
How they calculated the 70k number is completely irrelevant. The 70k number comes up only in making the the point that the base of existing applications for Windows presents an impenetrable barrier to market entry. No rational observer, from the Cato Institute or elsewhere, could possibly dispute this. Even Microsoft didn't, apparently. If the Cato estimate claimed there were only 10 Windows applications out there, well they might be right, but they would also have to admit there are only 11 applications total. Arguing about this number is just a bizarre exercise in miscommunication.
Calling these time-limited peep shows books is an egregious abuse of the word "book." I think it's only fair to doubt the honesty, competence, and judgement of any author, publisher, or med school who facilitates this kind of product. Producing these peep shows instead of real medical texts is not consistent with any legitimate educational purpose.
I guess it would be nice to know if stealing a domain name is theft or fraud, until legislation catches up with technology. But it seems like a focus on phrases like "property" and "designation for a service" is unlikely to lead to useful legislation. Whether we call domain names property or not, presumably the core issue is the entitlement of the registered owner of a domain name. It seems like the right answer is probably that domain names are in some ways like property (we'd like the right to buy/sell them) and in other ways like a designation for a service (somebody agrees to resolve your address). Ultimately, wouldn't it would be nice if the applicable laws had to do with the rights explicitly held by the domain name registrant, and not with the status of domain names as property or as designation for a service?
As a card carrying NAL (as in IANAL), it also seems to me like an important issue is whether or not the services provided by NSI (and ICANN?) should be considered a public accomodation (is that the right phrase?). If not, then presumably domain names entail roughly whatever rights they specify.
Regarding that one line, it seems like there are only a few possibilities here. The two most obvious to us non-lawyer types are that the authors of the memorandum actually don't know what "open source" means, or they're trying to mislead someone. But is it possibly just a lawyerly discourse thing, not to be taken seriously?
The rest of the document contains some mildly interesting stuff, including a note that "such reverse engineering is illegal." Is this true? My understanding from reading summaries of the DMCA (again, not a lawyer) is that reverse engineering is specifically permitted. But I also thought that DeCSS was built via means that are not well-described as reverse engineering. Are they just using the phrase in a different sense than the DMCA sense?
Lastly, I couldn't read this document without giving more thought to whether or not DeCSS facilitates "piracy." Although a lot of people have pointed out that cracking CSS is not the most straightforward way to copy DVDs, it seems like DeCSS can still be said to facilitate copying movies for distribution. If I have a computer with a DVD player and a fast internet connection, and I don't feel like doing a lot of hacking, it seems like the path of least resistance. But what's the issue here? Lots of things that aren't illegal facilitate things that are (e.g., guns, computers, ziploc(tm) bags). How is it relevant that DeCSS can be used for nefarious purposes? (I know this keeps coming up, but I still have no idea where they're going with this.)
It seems like abusing publicly bestowed trust in this way is itself an "improper purpose." Perhaps some clever lawyer could compel network solutions to take back its own domain name.
Many scientists keep electronic copies of articles they consider valuable or important in their field. (Not me, of course, I would never do this, unless it turns out it's not illegal.) I'm sure just about everyone I know keeps a little collection of pdf files, and collections of other kinds of data will certainly become popular to the extent that valuable articles are published in those formats.
Anyway, it seems to me this ought to produce a natural survival of the fittest articles. Those articles that are most widely appreciated will be cached in the most locations. In a large enough field, this is almost certainly already the case. If less popular articles (i.e., those least appreciated by the scientific masses) are lost, this is probably no greater a tragedy than the loss of work which goes unpublished (i.e., those least appreciated by 2-3 reviewers).
Of course, this enormous and rapidly growing archive is completely unorganized, and doesn't provide an easy mechanism for public access for the distant future. But the level of concern that valuable articles will be lost should be less than if people weren't already making enormous personal archives.
If I want maximally timely information with the most superficial possible coverage, news portals are the ideal medium. If I'm willing to wait a bit for slightly more analytical treatment, newspapers seem to fill the gap (I'm including electronic newspapers here -- I think the delivery medium is clearly a red herring). If I want the most analytical possible coverage, I will wait for someone to have pondered deeply enough about the subject to have written a book, journal article, or some other extended treatment, which can take weeks, months, or years.
So based on this, do newspapers have a future? If by newspapers we mean media that come out in discrete "editions" as opposed to continuously, then the question boils down to whether or not they provide some value for their more drawn out time scale. My impression is that most newspapers do not take adequate advantage of their comparatively leisurely time scale. Or, rather, most continuous news sources are able, by manpower or whatever other means, to provide coverage that is as good as most newspapers. There are exceptions (I like the NYT), but I can't see any purpose for newspapers that ask us to trade timeliness for no particular advantage.
I think this applies equally well to local news, even though most localities are as of the moment still best served by newspapers.
After reading yet another softball-filled interview (well, to be fair it looks like the reporter tried to ask tough questions but didn't follow through), I wonder if any reporters tried to ask Valenti the questions which come up over and over again on Slashdot? I haven't read every last Valenti interview available on this subject, but it seems like no one ever points out to him that DeCSS doesn't do much to prevent illegal redistribution. In the Salon interview, Damien Cave gets Valenti to say that any use of a DVD by which the consumer pays a fair price is fair use. But Cave doesn't follow up and ask Valenti to map this onto DeCSS. Valenti says he wholeheartedly endorses the prosecution of Jon Johansen, but the reporter never gets him to say why. We know from other interviews that Valenti for some reason thinks DeCSS will facilitate copying, but this reporter misses the opportunity to press him on that bizarre issue. Valenti then accuses his email detractors of illogic, while in the same response to a direct question about permitted uses of DVDs, all he really talks about is copying. The issue raised by DeCSS isn't copying, it's viewing. But both Valenti and Cave ignore this completely.
At the same time, at some level I sympathize with Valenti. He is trying to protect a very big industry, and he either knows or suspects that what he really wants is impossible. If I can see it, I can copy it. No copy protection scheme is liable to have any lasting effect on this, once they let me bring a copy into my home. I don't personally copy movies, and I don't plan to put them up on a server. But people will, and it's possible the movie industry will eventually be harmed as a result. Of course, that will take some time. Even music is not quite there yet, although MP3s sound reasonably good. But even if Valenti's concern is legitimate, I have a hard time mourning the Hollywood machine. As much as I have enjoyed the occasional big budget movie, I think they're a reasonable price to pay.
Slashdot Mirror
The author of a document has certain rights that are described in copyright law. For example, the right to produce derivative works or the right to distribute copies. As Lessig notes in the interview, consumers can also enter into agreements with copyright holders via things like shrink-wrap licenses. Entering into a restrictive agreement with someone is not tantamount to surrendering the rights an author has as copyright holder. Although it might strike some chord of fairness, copyright holders don't lose these rights just because you agreed not to read their books out loud (or whatever), any more than they lose their rights when they agree to let a publisher produce and sell copies of their work.
Just because the GPL doesn't grant certain privileges, that's no reason you can't obtain them anyway. Ask the code's author.
The existence of the GPL makes it easy for people to license their code under a certain set of terms. Just because those terms don't include the terms you want, that doesn't make it infectious. It's only an infection if you're too lazy to ask the author of the code for the permissions you want (that aren't included in the GPL), or if authors who are too lazy to answer their email fail to give permission for uses they would have deemed acceptable. If the author of the code would never say no to such a request, then sure, they should have released their code under a more permissive license. If they ever say no, then the GPL is actually protecting something they consider important. If the author is unreachable, then you have to accept their choice of the GPL at face value -- those are exactly the permissions they intended to grant.
The GPL only seems to infect everything it touches. If you want to use GPLed code but can't be GPL-compatible, nothing prevents you from contacting the author of the code for permission. Well, at least for smaller projects with easily reachable authors. For code that has hundreds of authors, some deceased, this might be more difficult.
dan
Indeed, this is huge. With this kernel, I believe NFS (esp NFSv3) moves from something you either have to pray your distribution vendor got right, or something you have to cobble together from patches, to a snazzy and high-functioning part of the Linux kernel right out of the box. Many thanks are owed to Trond Myklebust, David Higgen, Alan Cox, and the rest of the NFS gang.
I think the new Dune miniseries clearly owes more to the old Batman TV series than to Frank Herbert's or David Lynch's work. Although there have been a few parts that seemed well-dramatized, for the most part the zany costumes and comically exaggerated overacting (from both accomplished actors and unknowns, so I have to assume it was a decision of the director) scream camp. Although the novel has its flaws, and sure, in places it lends itself to a campy adaptation (esp. the earlier scenes with the Baron), the miniseries looks like it will pass on the opportunity to draw from the depth of the novel.
I suppose I am obligated to indicate how many times I've read Dune and the rest of the series. I have no idea though. Probably 30 for Dune, average of 5 for the rest.
dan
ps my subject line is not intended as a reference to the use of the word "batman" to refer to the character Patrin who shows up in heretics of dune.
The article about the Natapoff argument on the desirability of the electoral college system was interesting. It was a little light on details, but he seems to argue from the point of view that the only desideratum in designing an election system is maximizing individual power in voting. This seems like an assumption, not a research conclusion, and I wonder how widely it's shared. In light of the fact that Natapoff's quoted comments are, to say the least, deeply condescending to basically everyone, this seems like an important point. If he's going to claim to have proved the electoral college system is better than going by popular vote, at a minimum it seems like he ought to have a definition of "better" that is universally acceptable. Is the complete work available online anywhere?
Incidentally, it seems a little suspect to argue that democracy hinges on this system, when in the majority of elections it makes no difference whatsoever in the outcome. Correctly or not, people seem to feel they would have more voting power if votes for the second or third place candidate in your state could actually make a difference.
You are mistaken that patents do not spur research. If there were no patent protection, there would be no big drug companies. Drug companies sink an astronomical amount of money into obtaining and defending patents, and practically every day some drug company's stock shoots up or bottoms out because a patent is either obtained, upheld or shot down. We can argue about how good a job these drug companies do at developing drugs. We might even speculate about unethical machinations behind the "research" used to gain approval. But there's no question that patent protection underlies the entire industry. You don't sink billions of dollars into developing zoloft if it's going to sit between a generic equivalent and a better-marketed competitor's version on the shelf. And you can't keep a drug's composition a trade secret.
It's hardly different with copyright. For the time being, books are put out by publishers, and publishers stay in business because of copyright protection. A few publishers try to make money producing new editions of work that has run out of copyright protection. But it's pretty obvious from the distribution of books in bookstores that this is not where the money is. I think it's safe to say that Steven King's publishers would be a lot less thrilled about printing up millions of copies of his books if they knew that there would be competing editions within a week. And of course, say goodbye to hardcover editions.
I do have some personal interest in copyright protection, since I'm self-publishing a book. If there were no copyright protection, someone with more money than I have could easily produce a more attractive edition and sell it for less. But I still don't know if I really think copyright protection is a good thing. It runs against all the pure motivations for writing a book, and in my opinion there are too many people out there writing books for the wrong reasons.
In the case of patents, though, despite my reservations about the ethics, let alone the integrity, of the big drug companies, I think some patent protection is valuable. There are things I think we as people should want to have, that we can have a lot sooner if we compensate inventors with patent protection. Although it's true that coming up with a brilliant idea second is a real problem, this should rarely happen. Patents are supposed to apply to non-obvious inventions. Alas, the USPTO doesn't always see it that way. This doesn't mean that patent protection isn't valuable or desirable, it only means that it needs some tweaking.
dan
I wanted to order the CmdrTaco version, with all those DVD's, MP3's, Simpsons and South Park episodes pre-loaded, but I couldn't find the ordering page on CmdrTaco.net. Please advise.
It's not normally my inclination to try to re-allocate other people's efforts. After all, as much as the world needs biomedical engineers, it would be a shame if there weren't just a few people who devoted their lives to figuring out where all my socks go after I put them in the washer.
That said, some projects just scream out, "I have lots of energy and resources but I couldn't figure out anything useful to do with them." The web, for example. And possibly pi digit identification. So wouldn't it be nice to start a project to help herd people with that kind of muscle into projects they and everyone else would find more useful? Far be it from me to suggest a particular project that should have been diverted (for all I know this pi thing involved new frontiers in computational methods). But I'm sure there are lots of projects out there, in many fields, that haven't gotten done because nobody had both the interest and the resources.
I'd consider hosting such a project's web pages myself, I'm sure they would be extremely low-volume.
Not being one to trust anyone on anything, a few years ago, I (along with a colleague) decided to embark on a little experiment to see just how easy it was to render a floppy unreadable. We were concerned about everyday risks (and this guy happened to be doing some research that involved small electric motors). So we tried leaving them on top of monitors for a few minutes, leaving them on top of speakers, and moving the magnet end of the motors across the surface of the disk at various angles. There was one small Word file on this disk (on a Mac filesystem), and we used the same disk throughout. Despite repeated trials, especially with the otherwise powerful (in the paperclip sense) motor magnets, we were completely unable to erase the file or damage the filesystem.
I believe at the time our scientific conclusion was that floppies are not based on magnetism, but on "tiny bubbles of ectoplasmic phlogiston." We never tried the condition where the file was the sole copy of some critical document, I don't know if that would have affected the results.
"Implementing Windows emulation on Linux" isn't part of the plex86 picture. Like VMWare it allows you to run multiple operating systems simultaneously on one machine. Or rather two machines, one real one virtual. You still need to have Windows installed, you just don't have to reboot to use it.
I can imagine a few reasons why software like Plex86 or VMware might ultimately harm Microsoft. For example, they might help ease a Windows-to-Linux transition. But they don't let you run Windows software without Windows. Even Wine doesn't really do that (yet).
The RSA algorithm has been public knowledge since it was developed. It's release (a few weeks in advance of the patent expiration) simply means people can use it without a license from RSA.
RSA has not been cracked. Some specific RSA keys of particular lengths (e.g., 512 bits) have been discovered. That's no big deal, since we already know roughly how much computational power it should take to crack a given key. And some weaknesses in particular implementations of RSA have been noted. But it's reasonably well understood how much (implementation-independent) security is provided by a given key length, and notwithstanding advances in factoring, that has stood up pretty well.
RSA may or may not have something better but top secret up their sleeves, but if so it hasn't been exposed to the scrutiny of the RSA algorithm. And the most likely areas for improvement are in computational efficiency and things like that, not in security per se.
"HaX0rz are out there and eventually they will get in one way or another...Is that hopeless?"
Well, if you don't mind a little freeform juxtaposition, then yes! The real question is whether or not it's excessively hopeless, and I think the answer is yes there too. Security isn't easy, but it's possible to restrict risk to acceptable levels. I agree that backups and insurance are also important. But anyone who doesn't have these already knows they're incompetent.
And the insurance really shouldn't be so expensive for anyone who is conscientious about risk. To put this a little differently, I wonder how an insurance policy would sell if it had two tiers of coverage, the lower of which kicked in if it were found that at the time of the incident a relevant patch more than five days old was available. Well-run organizations would be happy to buy it, poorly run ones would have to stick with the expensive stuff.
Bruce Schneier seems like a pretty conscientious guy in print. But this article just reads like a detailed ad four counterpane's services.
In connection with his new book (which I haven't read yet, because I'm still trying to find a good consultant to find me a morally upstanding bookseller), I wonder how much of his attitude is a necessary contingency of running a security business, or if that's why he started counterpane in the first place. I don't find fault with his presentation of facts, more with the sense of hopelessness he has conveyed in recent writing (I'm going mostly by articles, excerpts, and his crypto-gram newsletter).
It seems like the problem to be solved here is, in reduced form, telling the difference between pages that are actually more relevant to the search criteria than others and pages that are trying to fake being especially relevant. That sounds like an AI-complete problem (actually it sounds a whole lot like a web page Turing test). In that case obscurity might be the best model, since the best solutions will probably be exploitable. (Actually, even real intelligence is probably exploitable.)
Zappa's Law (as relayed by StoryMan, who admits he may have gotten it wrong) may be cute, but it will get you into some serious trouble if you actually believe it in some consequential context. Let me give two pertinent examples.
.02%, but puts your company out of business when it gets posted on slashdot), it's nonetheless true that price shenanigans, heavy graphics, and all the stuff you and i may hate, have some connection with profits, and it's going to be hard to convince Amazon to stop trying to figure out what that connection is.
It's true that with respect to dying on a given flight, there are two major outcomes: dying and not. However, those two outcomes don't come in even proportions -- they aren't 50-50. If people from my country always fly on 50-50 Air, and people from your country fly on 99.99999-.00001 Air (both pretty shoddy organizations), then after a short period of time my country will be depopulated, while yours will thriving (but perhaps a bit nervous).
Now let's look at two online booksellers. 50-50 books figures either a book sells or it doesn't. PriceSetter Books is interested in the factors that contribute to the probability of a sale. PriceSetter figures out it can double it's sales of "Lucky Numbers" by changing the price from $13 to $12.95, and at the same time sell the same number of copies of "Worthless at Any Price" at $15 as it did at $12. 50-50 figures no matter what the price is, there are still exactly two outcomes, and just charges whatever they think is a good price. In both cases, a given consumer either buys the book or doesn't. But PriceSetter will earn more money, because they know that when you're talking about a million customers, there is a huge difference between 3% and 3.1%. And neither is all that close to 50-50.
Nobody really expects that booksellers won't try to maximize their profits by setting prices appropriately, whether profit is maximized in the short term, long term, or whatever (selling a book at a loss could maximize profits in some cases, I'm sure). But obviously in this case, people aren't happy with how Amazon went about it. I'm not so happy about it either, for many of the reasons listed in this thread.
Anyway, relevant to your message, I think we're stuck with the ad men and marketing slimeballs, because even if they're completely incompetent and come up with the wrong answers (price testing increases profits by
It seems like the only thing they could possibly be referring to as their intellectual property here is their specific (and earth-shatteringly trivial) encoding of the bar codes. There is certainly nothing innovative about the encoding technique. So it seems roughly equivalent to claiming a particular number as your intellectual property. I'd like to read the patent, but I'm already at capacity for patent-related outrage this week.
On the other hand, the bar code reader aside (which I've actually found helpful in testing some bar codes I was printing), I think the digital convergence stuff (meaning mainly the airwaves signal stuff) is kind of neat. It seems like a primitive but positive step towards some kind of actual (for lack of a better word) convergence. Not that I want any right now, but I still think it's cool.
(Also, is it true that you have to defend a patent or risk it falling into the public domain? I thought that was only true of trademarks.)
How they calculated the 70k number is completely irrelevant. The 70k number comes up only in making the the point that the base of existing applications for Windows presents an impenetrable barrier to market entry. No rational observer, from the Cato Institute or elsewhere, could possibly dispute this. Even Microsoft didn't, apparently. If the Cato estimate claimed there were only 10 Windows applications out there, well they might be right, but they would also have to admit there are only 11 applications total. Arguing about this number is just a bizarre exercise in miscommunication.
Calling these time-limited peep shows books is an egregious abuse of the word "book." I think it's only fair to doubt the honesty, competence, and judgement of any author, publisher, or med school who facilitates this kind of product. Producing these peep shows instead of real medical texts is not consistent with any legitimate educational purpose.
I guess it would be nice to know if stealing a domain name is theft or fraud, until legislation catches up with technology. But it seems like a focus on phrases like "property" and "designation for a service" is unlikely to lead to useful legislation. Whether we call domain names property or not, presumably the core issue is the entitlement of the registered owner of a domain name. It seems like the right answer is probably that domain names are in some ways like property (we'd like the right to buy/sell them) and in other ways like a designation for a service (somebody agrees to resolve your address). Ultimately, wouldn't it would be nice if the applicable laws had to do with the rights explicitly held by the domain name registrant, and not with the status of domain names as property or as designation for a service?
As a card carrying NAL (as in IANAL), it also seems to me like an important issue is whether or not the services provided by NSI (and ICANN?) should be considered a public accomodation (is that the right phrase?). If not, then presumably domain names entail roughly whatever rights they specify.
Regarding that one line, it seems like there are only a few possibilities here. The two most obvious to us non-lawyer types are that the authors of the memorandum actually don't know what "open source" means, or they're trying to mislead someone. But is it possibly just a lawyerly discourse thing, not to be taken seriously?
The rest of the document contains some mildly interesting stuff, including a note that "such reverse engineering is illegal." Is this true? My understanding from reading summaries of the DMCA (again, not a lawyer) is that reverse engineering is specifically permitted. But I also thought that DeCSS was built via means that are not well-described as reverse engineering. Are they just using the phrase in a different sense than the DMCA sense?
Lastly, I couldn't read this document without giving more thought to whether or not DeCSS facilitates "piracy." Although a lot of people have pointed out that cracking CSS is not the most straightforward way to copy DVDs, it seems like DeCSS can still be said to facilitate copying movies for distribution. If I have a computer with a DVD player and a fast internet connection, and I don't feel like doing a lot of hacking, it seems like the path of least resistance. But what's the issue here? Lots of things that aren't illegal facilitate things that are (e.g., guns, computers, ziploc(tm) bags). How is it relevant that DeCSS can be used for nefarious purposes? (I know this keeps coming up, but I still have no idea where they're going with this.)
It seems like abusing publicly bestowed trust in this way is itself an "improper purpose." Perhaps some clever lawyer could compel network solutions to take back its own domain name.
Many scientists keep electronic copies of articles they consider valuable or important in their field. (Not me, of course, I would never do this, unless it turns out it's not illegal.) I'm sure just about everyone I know keeps a little collection of pdf files, and collections of other kinds of data will certainly become popular to the extent that valuable articles are published in those formats.
Anyway, it seems to me this ought to produce a natural survival of the fittest articles. Those articles that are most widely appreciated will be cached in the most locations. In a large enough field, this is almost certainly already the case. If less popular articles (i.e., those least appreciated by the scientific masses) are lost, this is probably no greater a tragedy than the loss of work which goes unpublished (i.e., those least appreciated by 2-3 reviewers).
Of course, this enormous and rapidly growing archive is completely unorganized, and doesn't provide an easy mechanism for public access for the distant future. But the level of concern that valuable articles will be lost should be less than if people weren't already making enormous personal archives.
If I want maximally timely information with the most superficial possible coverage, news portals are the ideal medium. If I'm willing to wait a bit for slightly more analytical treatment, newspapers seem to fill the gap (I'm including electronic newspapers here -- I think the delivery medium is clearly a red herring). If I want the most analytical possible coverage, I will wait for someone to have pondered deeply enough about the subject to have written a book, journal article, or some other extended treatment, which can take weeks, months, or years.
So based on this, do newspapers have a future? If by newspapers we mean media that come out in discrete "editions" as opposed to continuously, then the question boils down to whether or not they provide some value for their more drawn out time scale. My impression is that most newspapers do not take adequate advantage of their comparatively leisurely time scale. Or, rather, most continuous news sources are able, by manpower or whatever other means, to provide coverage that is as good as most newspapers. There are exceptions (I like the NYT), but I can't see any purpose for newspapers that ask us to trade timeliness for no particular advantage.
I think this applies equally well to local news, even though most localities are as of the moment still best served by newspapers.
After reading yet another softball-filled interview (well, to be fair it looks like the reporter tried to ask tough questions but didn't follow through), I wonder if any reporters tried to ask Valenti the questions which come up over and over again on Slashdot? I haven't read every last Valenti interview available on this subject, but it seems like no one ever points out to him that DeCSS doesn't do much to prevent illegal redistribution. In the Salon interview, Damien Cave gets Valenti to say that any use of a DVD by which the consumer pays a fair price is fair use. But Cave doesn't follow up and ask Valenti to map this onto DeCSS. Valenti says he wholeheartedly endorses the prosecution of Jon Johansen, but the reporter never gets him to say why. We know from other interviews that Valenti for some reason thinks DeCSS will facilitate copying, but this reporter misses the opportunity to press him on that bizarre issue. Valenti then accuses his email detractors of illogic, while in the same response to a direct question about permitted uses of DVDs, all he really talks about is copying. The issue raised by DeCSS isn't copying, it's viewing. But both Valenti and Cave ignore this completely.
At the same time, at some level I sympathize with Valenti. He is trying to protect a very big industry, and he either knows or suspects that what he really wants is impossible. If I can see it, I can copy it. No copy protection scheme is liable to have any lasting effect on this, once they let me bring a copy into my home. I don't personally copy movies, and I don't plan to put them up on a server. But people will, and it's possible the movie industry will eventually be harmed as a result. Of course, that will take some time. Even music is not quite there yet, although MP3s sound reasonably good. But even if Valenti's concern is legitimate, I have a hard time mourning the Hollywood machine. As much as I have enjoyed the occasional big budget movie, I think they're a reasonable price to pay.