One was my girlfriend's system, the only one with out-of-date virus protection and, of course, the only unprotected machine with truly irreplaceable files.
You, a Linux user, willfully keep irreplaceable data on a Windows box? Perhaps you should rethink that practice.
The Linux boxes, of course, didn't even raise an eyebrow...
Maybe we could use this as a Linux success story to the unwashed masses?
Actually, this is done quite nicely by Java. And just think, it's had this feature for like 7 years now (ie, since day one)...
Show me a piece of Java code that can get through the Java sandbox without exploiting a bug in a particular implementation (ie, by exploiting a loophole in the security model itself).
Most in game browsers are really half assed and lame as hell.
The exception to this is the in-game browser in Half-Life, which has all the functionality of Arcade (most notably SmartSpy style query limiting) without any of the crap.
Unless I'm mistaken, Quake 3 and offspring also have this.
Launching the appropriate game (determined by URI scheme, I assume) and instructing it to attempt to connect to a game server is not executing arbitrary code. It's just like using a common server browser, only this time, you can (maybe) use a Web browser to browse what games are running on the machine, to have a hyperlink to a running game on a Web page that you can just click on to connect to, etc.
As you may already know, Unreal also uses URLs in the way you describe. It's pretty damn slick too. It also opens the possibility to a single server serving multiple games in (different threads of) the same process. Imagine:
doom3://doom3.idsoftware.com/ctf/game1
doom3://doom3.idsoftware.com/dm/slaughterhouse
...etc...
By the way, if you use Java any, I suspect it would make implementation of such a program much easier, thanks to various facilities for manipulating URIs, the serialization system, easy-to-use network APIs, etc. Maybe make a companion "universal master server" program for it, too. Oh, and of course, this would be quite portable.
That's not logical at all. Since the compiler already knows that you're instantiating B and not A, it already has enough information to set up the vtables appropriately. Why does it need to call the constructor first? IMNSHO, Java gets it right -- the vtables are set up well before the constructor gets called, so calling methods (which are always virtual, unless they are final) always yields the expected (and most useful) results.
Actually, upon further observation, it would seem that even GCC has this behavior! Shame, shame. What can I say; I'm jaded from my Java experience, where all methods are virtual, and the vtables are set up well before the constructor gets invoked.
a development manager who can't understand why you don't call virtuals from a constructor
Since when does one not call virtuals from a constructor? Unless your compiler really sucks and the generated code doesn't set up the vtables before passing control to the constructor (which, in case I forgot to mention it, means your compiler really sucks), it should be fine.
So, I can post a little Terms of Service link on the bottom of my Web site that says that, by viewing this site, you acknowledge and agree that I may gun you down in broad daylight in the middle of Times Square for any reason? And expect to get away with it? I think not!
That's not even an EULA. It's just this little document they hope you'll read, which they call the "Terms of Service". Supposedly, by using the site, you agree to abide by the TOS, but since you've already used the site (you had to hit one of their pages to see the link to the TOS, and viewing the TOS also constitutes use of the site), their TOS isn't worth the pixels it's displayed on, since you can't agree to a contract before you've even read it! And that's assuming the TOS is valid at all, which is highly questionable.
Your Republican philosophy is all fine and good, and I find it somewhat agreeable, but remember that the assholes you vote for (*cough*Dubya*cough*) are getting paid off left and right, and don't give a damn about you or any of the rest of their so-called "constituents". All they care about is lining their pockets through "campaign contributions" (*cough*personal*cough*favors*cough*).
Since you seem intelligent enough to do so, I suggest you think about this at the next election. Or tell me why I'm wrong, as I really wish I could be more confident in my president.
Spam already is outlawed. The problem is that the law is unenforceable (anyone in their right mind knows the mandatory 'remove' address just causes more spam to come your way), but it is, nevertheless, on the books.
People will get trampled enough to realize what's happening. Even you acknowledge that. When that happens, whether or not there's a legal way regain control is really irrelevant. It wasn't relevant in the American Revolution (just saying bad things about England was treason back then, let alone actively pushing back), and it sure as hell won't be relevant now.
While today's public schools do indeed teach kids to sit down and do what they're told, this is pretty much what was taught at the time of the American Revolution. Go crack a history book and see how much good it did them. It's even less effective today, with so many kids focusing almost single-mindedly on not doing what they're told.
As for protests, I don't think the problem is that the government can't see the difference between them and riots. The problem is that the government doesn't want to hear what the protestors have to say, for obvious reasons involving being paid not to listen (since big business is their constituency, like in the 1900s). That's why they take those new and creative to deter protestors, such as forcing them a few miles away from the site of whatever it is they're protesting and arresting anyone who refuses. Again, this doesn't matter. Gandhi and his followers got arrested left and right, despite protesting peacefully, and guess what? They won anyway!
The reason the dividing line between the government and Microsoft is growing fainter is that George W. Bush is at the helm. Why else do you think the DoJ dropped their case against MS shortly after he took office? It follows that, once he is removed, Microsoft will be in for another ass-kicking. Besides, those nine dissenting states are also quite capable of putting them in a world of hurt, and that's exactly what they seem intent on doing.
As for Microsoft negotiating with the governments of foreign countries on a high-level basis, look what it got 'em in Peru -- an intense backlash from a Peruvian congressman! And no, it's not just in Peru. Many European governments are also realizing how dangerous Microsoft can and will be.
It'll probably be legal to remove these bandits from power once the time is right. If it isn't, tough; they'll be removed anyway.
Current-day practice is not to have a "tear-open" agreement, but, instead, the agreement is presented when the user attempts to install the software. The user had no knowledge of the agreement's existence, let alone its terms, when the user paid for the software. By this lawyer's logic, that makes the agreement null and void.
Click-wrap licenses usually tell the reader to return the software to its place of purchase for a full refund, if the user refuses to be bound by it. Unfortunately, the place of purchase will generally not take it back or refund the user, as an understandable matter of policy (they have no way of knowing if you copied the distribution media prior to returning the software). However, this effectively means that the user is forced to either accept the terms of the agreement, or not use the software and let it rot, since they can't get it refunded.
Often, click-wrap licenses state that opening the package constitutes acceptance. However, you didn't even see the license until you attempted to install the software (which obviously happens after opening the package).
One big point you're missing here is that this is a patent, not a law. You're still quite free to install a non-DRM OS on your computer (and use a kernel debugger, if you so desire). The CBDTPA would of course change that, but with all the yelling from the EFF and even big companies, it's as good as dead already (unless of course everyone quits yelling -- this is not an invitation to become complacent!).
Also, this patent does not say anything about external enforcement (ie, by the law) of not being allowed to run a kernel debugger. It simply says that the OS should either prevent the kernel debugger from running, or else discard its trustworthy identity (which is akin to voiding the warranty on some device when you open it up and tinker with it).
Of course, this is a potential slippery slope scenario, and I share your concern for that reason. But in and of itself, this patent is not much of a threat.
Security and liability. Somewhere down the line, security holes in M$ software started costing consumers and companies millions, perhaps billions. The developers of said software should bear some legal responsibility to make secure code. If they don't then there is no incentive for M$ to even fix the bugs in a timely manner.
Actually, the NIST has something to say about that too, as reported in this article.
Why would a manufacturer want to take the chance that they'd be involved in a messy class-action suit from customers because their portable media player now can't play music?
Why would a manufacturer even stop to think about that possibility? They wouldn't.
If they ever lent even one brain cycle to long-terms risks such as this, they would already be making statements against DRM (since that makes their customers happier, and customer unhappiness is a long-term risk for obvious reasons).
Actually the legality is the only problem here. Any technical measures used to enforce the EULA are just cracker fodder. They WILL be broken, and in short order. If you don't believe me, go look at the evidence.
Actually, that just means you get to choose which one you're bound to. Since you seem to prefer the terms of the Windows Update version, go to Windows Update and download the patch from there.
Then you're not bound by it, since you haven't agreed to it. Of course, you also have to prove you didn't agree to it in a court of law, and Microsoft will have quietly "fixed the bug" behind the scenes in a new version of the auto-updater (and you can only download the latest version), thus "proving" to the court that you did agree to the EULA, even though you actually didn't since you were running an older version of the auto-updater, but you can't demonstrate to the court that the older version doesn't show you the EULA, because it's already "upgraded" itself (or, if it hasn't, perhaps because you disabled it, they've used their newly acquired root access to "upgrade" it for you).
You could try making a backup copy of your disk drive(s) and restore them to present to the court, but Microsoft will surely allege that the backups were altered to disable the auto-updater's prompting, and you have no real way of proving their authenticity.
Altering the updater this way is a violation of the DMCA, by the way -- it constitutes circumvention. It's also a breach of the EULA you agreed to when you downloaded the auto-updater. Of course, you did no such thing, and therefore did not violate the DMCA nor breach the auto-updater's EULA, but you can't prove that, and since you're guilty until proven innocent, do not pass go, do not collect $200, proceed directly to jail.
And even if you could prove this to them, they would simply drag out the process until you can no longer pay your lawyers and court fees, thus turning a court battle into a game of last man standing, which you will lose, because Microsoft has far more money than you.
If you're going to challenge Microsoft, do it on a level playing field, or preferably on a playing field tipped in your favor. Widespread civil disobedience should do the trick nicely, but you need a lot of people for that. Since Microsoft will inevitably become intolerable even to Joe Clueless on the street, this will not be a problem in the future, so maybe you should just wait quietly until people start getting pissed off, then join the fray. Patience is an enormously powerful weapon; use it well.
Well, I would have used public-key crypto (SSL, etc) to perform authentication, so that I have free reign over the machine (since I have the 1024-byte secret key), but by the time anyone else cracks it, nobody would be running that OS anyway.
You're right, the scenarios you've described are unrealistic. History has a remarkable aptitude for repeating itself. As with antitrust in the 1900s and IBM in the 1970s and 1980s, so too will the mighty Microsoft empire fall. When Microsoft and similar companies gain too much foothold, it will be obvious even to Joe Clueless on the street that they need to be stopped. Once Joe Clueless on the street decides they need to be stopped, Microsoft's annihilation is only a matter of time.
Of course, that's if the government doesn't fusion-bomb us all, but then there's not much left to control, so I doubt they will.
Slashdot Mirror
Show me a piece of Java code that can get through the Java sandbox without exploiting a bug in a particular implementation (ie, by exploiting a loophole in the security model itself).
Unless I'm mistaken, Quake 3 and offspring also have this.
Launching the appropriate game (determined by URI scheme, I assume) and instructing it to attempt to connect to a game server is not executing arbitrary code. It's just like using a common server browser, only this time, you can (maybe) use a Web browser to browse what games are running on the machine, to have a hyperlink to a running game on a Web page that you can just click on to connect to, etc.
Since when did I value GameSpy's products enough to pay for them? I've never paid the bastards a dime!
By the way, if you use Java any, I suspect it would make implementation of such a program much easier, thanks to various facilities for manipulating URIs, the serialization system, easy-to-use network APIs, etc. Maybe make a companion "universal master server" program for it, too. Oh, and of course, this would be quite portable.
That's not logical at all. Since the compiler already knows that you're instantiating B and not A, it already has enough information to set up the vtables appropriately. Why does it need to call the constructor first? IMNSHO, Java gets it right -- the vtables are set up well before the constructor gets called, so calling methods (which are always virtual, unless they are final) always yields the expected (and most useful) results.
Actually, upon further observation, it would seem that even GCC has this behavior! Shame, shame. What can I say; I'm jaded from my Java experience, where all methods are virtual, and the vtables are set up well before the constructor gets invoked.
So, I can post a little Terms of Service link on the bottom of my Web site that says that, by viewing this site, you acknowledge and agree that I may gun you down in broad daylight in the middle of Times Square for any reason? And expect to get away with it? I think not!
That's not even an EULA. It's just this little document they hope you'll read, which they call the "Terms of Service". Supposedly, by using the site, you agree to abide by the TOS, but since you've already used the site (you had to hit one of their pages to see the link to the TOS, and viewing the TOS also constitutes use of the site), their TOS isn't worth the pixels it's displayed on, since you can't agree to a contract before you've even read it! And that's assuming the TOS is valid at all, which is highly questionable.
Since you seem intelligent enough to do so, I suggest you think about this at the next election. Or tell me why I'm wrong, as I really wish I could be more confident in my president.
Spam already is outlawed. The problem is that the law is unenforceable (anyone in their right mind knows the mandatory 'remove' address just causes more spam to come your way), but it is, nevertheless, on the books.
People will get trampled enough to realize what's happening. Even you acknowledge that. When that happens, whether or not there's a legal way regain control is really irrelevant. It wasn't relevant in the American Revolution (just saying bad things about England was treason back then, let alone actively pushing back), and it sure as hell won't be relevant now.
While today's public schools do indeed teach kids to sit down and do what they're told, this is pretty much what was taught at the time of the American Revolution. Go crack a history book and see how much good it did them. It's even less effective today, with so many kids focusing almost single-mindedly on not doing what they're told.
As for protests, I don't think the problem is that the government can't see the difference between them and riots. The problem is that the government doesn't want to hear what the protestors have to say, for obvious reasons involving being paid not to listen (since big business is their constituency, like in the 1900s). That's why they take those new and creative to deter protestors, such as forcing them a few miles away from the site of whatever it is they're protesting and arresting anyone who refuses. Again, this doesn't matter. Gandhi and his followers got arrested left and right, despite protesting peacefully, and guess what? They won anyway!
The reason the dividing line between the government and Microsoft is growing fainter is that George W. Bush is at the helm. Why else do you think the DoJ dropped their case against MS shortly after he took office? It follows that, once he is removed, Microsoft will be in for another ass-kicking. Besides, those nine dissenting states are also quite capable of putting them in a world of hurt, and that's exactly what they seem intent on doing.
As for Microsoft negotiating with the governments of foreign countries on a high-level basis, look what it got 'em in Peru -- an intense backlash from a Peruvian congressman! And no, it's not just in Peru. Many European governments are also realizing how dangerous Microsoft can and will be.
It'll probably be legal to remove these bandits from power once the time is right. If it isn't, tough; they'll be removed anyway.
Current-day practice is not to have a "tear-open" agreement, but, instead, the agreement is presented when the user attempts to install the software. The user had no knowledge of the agreement's existence, let alone its terms, when the user paid for the software. By this lawyer's logic, that makes the agreement null and void.
Click-wrap licenses usually tell the reader to return the software to its place of purchase for a full refund, if the user refuses to be bound by it. Unfortunately, the place of purchase will generally not take it back or refund the user, as an understandable matter of policy (they have no way of knowing if you copied the distribution media prior to returning the software). However, this effectively means that the user is forced to either accept the terms of the agreement, or not use the software and let it rot, since they can't get it refunded.
Often, click-wrap licenses state that opening the package constitutes acceptance. However, you didn't even see the license until you attempted to install the software (which obviously happens after opening the package).
Also, this patent does not say anything about external enforcement (ie, by the law) of not being allowed to run a kernel debugger. It simply says that the OS should either prevent the kernel debugger from running, or else discard its trustworthy identity (which is akin to voiding the warranty on some device when you open it up and tinker with it).
Of course, this is a potential slippery slope scenario, and I share your concern for that reason. But in and of itself, this patent is not much of a threat.
It's far better than Windows Media Player, Open Source or no, but it'd be nice...
If they ever lent even one brain cycle to long-terms risks such as this, they would already be making statements against DRM (since that makes their customers happier, and customer unhappiness is a long-term risk for obvious reasons).
Actually the legality is the only problem here. Any technical measures used to enforce the EULA are just cracker fodder. They WILL be broken, and in short order. If you don't believe me, go look at the evidence.
Actually, that just means you get to choose which one you're bound to. Since you seem to prefer the terms of the Windows Update version, go to Windows Update and download the patch from there.
Then you're not bound by it, since you haven't agreed to it. Of course, you also have to prove you didn't agree to it in a court of law, and Microsoft will have quietly "fixed the bug" behind the scenes in a new version of the auto-updater (and you can only download the latest version), thus "proving" to the court that you did agree to the EULA, even though you actually didn't since you were running an older version of the auto-updater, but you can't demonstrate to the court that the older version doesn't show you the EULA, because it's already "upgraded" itself (or, if it hasn't, perhaps because you disabled it, they've used their newly acquired root access to "upgrade" it for you).
You could try making a backup copy of your disk drive(s) and restore them to present to the court, but Microsoft will surely allege that the backups were altered to disable the auto-updater's prompting, and you have no real way of proving their authenticity.
Altering the updater this way is a violation of the DMCA, by the way -- it constitutes circumvention. It's also a breach of the EULA you agreed to when you downloaded the auto-updater. Of course, you did no such thing, and therefore did not violate the DMCA nor breach the auto-updater's EULA, but you can't prove that, and since you're guilty until proven innocent, do not pass go, do not collect $200, proceed directly to jail.
And even if you could prove this to them, they would simply drag out the process until you can no longer pay your lawyers and court fees, thus turning a court battle into a game of last man standing, which you will lose, because Microsoft has far more money than you.
If you're going to challenge Microsoft, do it on a level playing field, or preferably on a playing field tipped in your favor. Widespread civil disobedience should do the trick nicely, but you need a lot of people for that. Since Microsoft will inevitably become intolerable even to Joe Clueless on the street, this will not be a problem in the future, so maybe you should just wait quietly until people start getting pissed off, then join the fray. Patience is an enormously powerful weapon; use it well.
Well, I would have used public-key crypto (SSL, etc) to perform authentication, so that I have free reign over the machine (since I have the 1024-byte secret key), but by the time anyone else cracks it, nobody would be running that OS anyway.
Of course, that's if the government doesn't fusion-bomb us all, but then there's not much left to control, so I doubt they will.