IMO, it would be a serious bug in a program to have dependencies, but not install them at the same time as the program. Your argument's as spurious as a program that requires.NET but doesn't come with the.NET installer; no, that won't work either for someone who doesn't have it already. How is the situation different here?
Assuming a dependence on IE itself (as opposed to, say, Trident, or just an HTML renderer) is certainly plausible (there are such programs), but they're mostly going to be webapps and thus require the Internet anyway. For ones that don't, you're going to have to install the program somehow; so why not install IE the same way at the same time? (It's not as if it'd be hard to find copies; Microsoft would almost certainly have a copy on their website somewhere, and the company who distributes their IE-requiring software could find it and distribute it at the same time.) I just don't see why you seem to think IE would have a special status in all this; although it may be the most common web browser to use as a dependency, there are many other dependencies that programs have which are far more common, so this problem's been solved time and time again in the context of libraries that aren't web browsers.
Incidentally, I don't have Internet access at home, so don't assume I don't know that computers don't need to be connected to the Internet. (Thus Slashdotting at work, lucky me.)
I once used the old Live Search in order to grab a webpage from its cache which wasn't in Google's cache for some reason. I suspect that's rather a corner case, though.
I generally use Wikipedia to search for a company's website, rather than Google; it may be inaccurate for some things, but it's pretty much 100% accurate for that.
Now, why would Novell sign such an agreement? Easy: Because their legal department advised them to do so. From this we can conclude that Novells legal department has knowledge of legal risks concerning Mono.
Actually, I suspect they signed it because Microsoft offered them lots of money to do so. Imagine Microsoft offered to pay you $1000 and also give you a patent licence to Mono. Would you accept their offer? Would that then mean that you believed Mono required a patent licence to use?
Then why would you be trying to connect to it? The error message would be a lot more careful to explain that the problem was a lack of Internet, rather than the lack of a browser.
Also, I feel confident in predicting that 100% of people who have posted a comment on Slashdot have, or have had at least once, some method of accessing the Internet. Otherwise, how could they post the comments?
Err... on Slashdot of all places, I'd expect you to know that it's possible to connect to the Internet without a browser. Which is what the button would do, in order to download the browser.
Umm... except there's a button offered to do it for you, in that example. Computers are perfectly capable of doing downloads without a browser; it's just that the UI is substantially worse than what you're used to. (Windows comes with both ftp and telnet; and copies of wget aren't hard to get hold of on any operating system. None is a browser, yet all can download things from the Internet.) It isn't sensible to insist that people who know what they're doing know how to use ftp to get a browser from the Internet; it is sensible to expect them to be able to click on a button designed to do the download for them (in the absence of a browser), though.
It's worth mentioning that WINE plugs Mozilla into the other end of the APIs in question, rather than IE, so it's certainly doable. Of course, I'm not quite sure how well WINE's solution works compared to the native Windows one in this regard; I have seen at least one program work correctly doing this, but that's a long way from saying that every program does.
I'm having fun over here running Linux whilst complying with the letter of the connect-to-our-networks policy at this University, which was written assuming Windows (I comply with the spirit too, but in a different way). For instance, I have antivirus software installed, just because the rules say that I have to; it doesn't have a whole lot to look for (there aren't many known viruses for Linux, so it mostly looks for Windows viruses on the offchance, and for other potentially problematic things such as logic bombs in zip files), but it's there. I get portscanned every time I connect, and yes, I do have iptables set up (via ufw) to drop those packets... (On the other hand, I'd drop those packets from anywhere else too; there are open ports but with random high numbers, none of the standard ones, and they're only open from certain directions. Also, with the exception of port 80, there isn't anything behind most of the standard ports to scan; the other daemons I only run when I plan to connect to them.) So it's not a case of "just to spite"; it's a case of "portscan clean, and enhance security at the same time".
Now to figure out why I need to use KDE to connect to the connection here, and Gnome at home, and why it's the desktop environment that makes a difference rather than the network management applet...
Err, it's above 1% even by Microsoft-sponsored estimates now. (And there was an article on Slashdot a while back showing an internal Microsoft slide that claimed Linux had a bigger market share than OS X, but I don't think anyone believes it, although it was mentioned that Microsoft probably have more accurate Linux adoption figures than anyone else.) Linux probably isn't big enough to dent Microsoft's profits excessively yet (except when Microsoft take actions that hurt themselves to also hurt Linux); however, it is big enough to show up on the radar.
Wait, I wrote all that, then noticed you were probably being sarcastic. However, I'll submit this comment anyway, just because I took all the trouble to write it...
Not quite. You need to go after people who can afford to settle with you for lots of money, but who aren't so rich that they can waste money running the entire court case through just to spite you (which is possibly what's happening in SCO vs IBM).
That sort of thing happens everywhere, not just Croatia. (The latest mobile phone pricing gimmick over here in the UK is to multiply all their prices by a factor, and money topped-up onto prepay phone tariffs by the same factor; so we have loads of adverts saying things like "top up £15 and get £50 worth of calls!". So we now have the weird situation in which you pay pounds to prepay on your phone, and you get credit for making calls in units which are called pounds, but aren't.)
You're correct, but for the sake of completeness, I'll add to that: there are also locations in the world (mostly islands substantially smaller than Great Britain or Ireland) that are part of the UK, but not of Great Britain nor anywhere in Island.
I have won NetHack without exploiting bugs. So have many other people.
Nowadays, it's reached a point where it's debatable whether most of the commonly-abused things in the game (such as pudding farming) are genuine bugs or not. (Also, most abuses are either sufficiently minor that they aren't worth it, or sufficiently boring that they aren't worth it.) There are a few genuinely abusable bugs (such as Astral Call - attempting to rename one of the priests on the final level and determining information you aren't meant to know from the error message), but they tend to be fixed on public servers.
If you can't solve NetHack without cheating, obviously you need to read more spoilers... (It is too difficult to really be able to solve unspoilt, without spending several months working out all the mechanics of the game by experiment.)
Except, that Windows supports multiple formats on the clipboard simultaneously. For instance, copying text might put both a plain-text version and a formatted version in HTML there; it might even put an OLE object there. The application that pastes from the clipboard can pick a format it understands from the list there. (If you want proof, copy some text from, say, Word, then go to Paste Special... in any Windows program that obeys the standard menu rules (i.e. not ribbon-based; this might work on ribbon-based programs too, but I'm not sure how). The list of formats there is a list of all the formats on the clipboard that the application you're pasting into understands.
This has been working since Windows 3.1; I'm not entirely sure how Microsoft managed to break it. (Unless it was a deliberate attempt to stop third-party programs copying/pasting correctly with Word, etc? I don't know.)
Not just speed up; if a DOS program runs for long enough (or maybe produces enough output, I'm not sure) without input, its output drops to the rate of 1 character per second, for some reason I don't understand. Pressing any modifier key (Alt, Control, Shift, or whatever) brings it back up to full speed for a while.
Are those the same NaN, or different NaNs, I wonder? (IEEE-format NaNs carry a "payload" which is some extra data; IIRC, it's intended for debug purposes, or something like that. In certain languages, you can see it if you print the NaN.)
Actually, some debug information formats can handle the lack of a frame pointer nowadays (the same way as the compiler, by keeping track of what the offests should be); I seem to remember that one of gcc's defaults (maybe for a particular optimisation setting?) omits the frame pointer if and only if there's a debug format available capable of debugging the program anyway. On the whole, though, you're correct.
I've updated from every version of Ubuntu since Feisty; that's Feisty->Gutsy, Gutsy->Hardy, Hardy->Intrepid, Intrepid->Jaunty beta 2 (->Jaunty, but that isn't a distro upgrade). Intrepid->Jaunty was the only one which actually went smoothly, but in two of the others it was my fault (one was a stray file with a significant name in/usr/local/lib, I don't know how I did that; the other was due to a hardware problem causing my computer to shut down in the middle of an upgrade, but nevertheless it was possible to resume the update (in text mode) once the computer came back up). So I don't think updating Ubuntu is necessarily doomed to failure; although it doesn't seem to be quite as good as certain other distributions.
I thought the general Slashdot opinion (I have no idea whether it's right or not, I'm not a lawyer and I live in the wrong country to know American law) was that although downloading meant that copyright infringement had been committed, it was the person you downloaded from that made the copy, not you.
Over here in the UK, when something like that happens, we generally tell the media and let them and the corporation fight it out. (You sometimes even get your money back when that happens.) Complaining at large companies makes good TV over here, so the TV companies are generally happy to investigate; and although things don't get all that much better, at least they don't get worse.
Slashdot Mirror
IMO, it would be a serious bug in a program to have dependencies, but not install them at the same time as the program. Your argument's as spurious as a program that requires .NET but doesn't come with the .NET installer; no, that won't work either for someone who doesn't have it already. How is the situation different here?
Assuming a dependence on IE itself (as opposed to, say, Trident, or just an HTML renderer) is certainly plausible (there are such programs), but they're mostly going to be webapps and thus require the Internet anyway. For ones that don't, you're going to have to install the program somehow; so why not install IE the same way at the same time? (It's not as if it'd be hard to find copies; Microsoft would almost certainly have a copy on their website somewhere, and the company who distributes their IE-requiring software could find it and distribute it at the same time.) I just don't see why you seem to think IE would have a special status in all this; although it may be the most common web browser to use as a dependency, there are many other dependencies that programs have which are far more common, so this problem's been solved time and time again in the context of libraries that aren't web browsers.
Incidentally, I don't have Internet access at home, so don't assume I don't know that computers don't need to be connected to the Internet. (Thus Slashdotting at work, lucky me.)
I once used the old Live Search in order to grab a webpage from its cache which wasn't in Google's cache for some reason. I suspect that's rather a corner case, though.
I generally use Wikipedia to search for a company's website, rather than Google; it may be inaccurate for some things, but it's pretty much 100% accurate for that.
Now, why would Novell sign such an agreement? Easy: Because their legal department advised them to do so. From this we can conclude that Novells legal department has knowledge of legal risks concerning Mono.
Actually, I suspect they signed it because Microsoft offered them lots of money to do so. Imagine Microsoft offered to pay you $1000 and also give you a patent licence to Mono. Would you accept their offer? Would that then mean that you believed Mono required a patent licence to use?
Then why would you be trying to connect to it? The error message would be a lot more careful to explain that the problem was a lack of Internet, rather than the lack of a browser. Also, I feel confident in predicting that 100% of people who have posted a comment on Slashdot have, or have had at least once, some method of accessing the Internet. Otherwise, how could they post the comments?
Err... on Slashdot of all places, I'd expect you to know that it's possible to connect to the Internet without a browser. Which is what the button would do, in order to download the browser.
That's not a particularly good experiment; the second post is arguably Redundant.
You have Mozilla's IP address memorized, but don't know the URL to their FTP site? Or were you just showing off?
Umm... except there's a button offered to do it for you, in that example. Computers are perfectly capable of doing downloads without a browser; it's just that the UI is substantially worse than what you're used to. (Windows comes with both ftp and telnet; and copies of wget aren't hard to get hold of on any operating system. None is a browser, yet all can download things from the Internet.) It isn't sensible to insist that people who know what they're doing know how to use ftp to get a browser from the Internet; it is sensible to expect them to be able to click on a button designed to do the download for them (in the absence of a browser), though.
It's worth mentioning that WINE plugs Mozilla into the other end of the APIs in question, rather than IE, so it's certainly doable. Of course, I'm not quite sure how well WINE's solution works compared to the native Windows one in this regard; I have seen at least one program work correctly doing this, but that's a long way from saying that every program does.
I'm having fun over here running Linux whilst complying with the letter of the connect-to-our-networks policy at this University, which was written assuming Windows (I comply with the spirit too, but in a different way). For instance, I have antivirus software installed, just because the rules say that I have to; it doesn't have a whole lot to look for (there aren't many known viruses for Linux, so it mostly looks for Windows viruses on the offchance, and for other potentially problematic things such as logic bombs in zip files), but it's there. I get portscanned every time I connect, and yes, I do have iptables set up (via ufw) to drop those packets... (On the other hand, I'd drop those packets from anywhere else too; there are open ports but with random high numbers, none of the standard ones, and they're only open from certain directions. Also, with the exception of port 80, there isn't anything behind most of the standard ports to scan; the other daemons I only run when I plan to connect to them.) So it's not a case of "just to spite"; it's a case of "portscan clean, and enhance security at the same time". Now to figure out why I need to use KDE to connect to the connection here, and Gnome at home, and why it's the desktop environment that makes a difference rather than the network management applet...
Usual defensive acquiring of domains, to stop anyone else registering them. Most major companies do it, there's nothing unusual involved there.
But to do that, you have to click twice! That's so old-fashioned...
Err, it's above 1% even by Microsoft-sponsored estimates now. (And there was an article on Slashdot a while back showing an internal Microsoft slide that claimed Linux had a bigger market share than OS X, but I don't think anyone believes it, although it was mentioned that Microsoft probably have more accurate Linux adoption figures than anyone else.) Linux probably isn't big enough to dent Microsoft's profits excessively yet (except when Microsoft take actions that hurt themselves to also hurt Linux); however, it is big enough to show up on the radar. Wait, I wrote all that, then noticed you were probably being sarcastic. However, I'll submit this comment anyway, just because I took all the trouble to write it...
Not quite. You need to go after people who can afford to settle with you for lots of money, but who aren't so rich that they can waste money running the entire court case through just to spite you (which is possibly what's happening in SCO vs IBM).
That sort of thing happens everywhere, not just Croatia. (The latest mobile phone pricing gimmick over here in the UK is to multiply all their prices by a factor, and money topped-up onto prepay phone tariffs by the same factor; so we have loads of adverts saying things like "top up £15 and get £50 worth of calls!". So we now have the weird situation in which you pay pounds to prepay on your phone, and you get credit for making calls in units which are called pounds, but aren't.)
You're correct, but for the sake of completeness, I'll add to that: there are also locations in the world (mostly islands substantially smaller than Great Britain or Ireland) that are part of the UK, but not of Great Britain nor anywhere in Island.
I have won NetHack without exploiting bugs. So have many other people. Nowadays, it's reached a point where it's debatable whether most of the commonly-abused things in the game (such as pudding farming) are genuine bugs or not. (Also, most abuses are either sufficiently minor that they aren't worth it, or sufficiently boring that they aren't worth it.) There are a few genuinely abusable bugs (such as Astral Call - attempting to rename one of the priests on the final level and determining information you aren't meant to know from the error message), but they tend to be fixed on public servers. If you can't solve NetHack without cheating, obviously you need to read more spoilers... (It is too difficult to really be able to solve unspoilt, without spending several months working out all the mechanics of the game by experiment.)
Except, that Windows supports multiple formats on the clipboard simultaneously. For instance, copying text might put both a plain-text version and a formatted version in HTML there; it might even put an OLE object there. The application that pastes from the clipboard can pick a format it understands from the list there. (If you want proof, copy some text from, say, Word, then go to Paste Special... in any Windows program that obeys the standard menu rules (i.e. not ribbon-based; this might work on ribbon-based programs too, but I'm not sure how). The list of formats there is a list of all the formats on the clipboard that the application you're pasting into understands. This has been working since Windows 3.1; I'm not entirely sure how Microsoft managed to break it. (Unless it was a deliberate attempt to stop third-party programs copying/pasting correctly with Word, etc? I don't know.)
Not just speed up; if a DOS program runs for long enough (or maybe produces enough output, I'm not sure) without input, its output drops to the rate of 1 character per second, for some reason I don't understand. Pressing any modifier key (Alt, Control, Shift, or whatever) brings it back up to full speed for a while.
Are those the same NaN, or different NaNs, I wonder? (IEEE-format NaNs carry a "payload" which is some extra data; IIRC, it's intended for debug purposes, or something like that. In certain languages, you can see it if you print the NaN.)
Actually, some debug information formats can handle the lack of a frame pointer nowadays (the same way as the compiler, by keeping track of what the offests should be); I seem to remember that one of gcc's defaults (maybe for a particular optimisation setting?) omits the frame pointer if and only if there's a debug format available capable of debugging the program anyway. On the whole, though, you're correct.
I've updated from every version of Ubuntu since Feisty; that's Feisty->Gutsy, Gutsy->Hardy, Hardy->Intrepid, Intrepid->Jaunty beta 2 (->Jaunty, but that isn't a distro upgrade). Intrepid->Jaunty was the only one which actually went smoothly, but in two of the others it was my fault (one was a stray file with a significant name in /usr/local/lib, I don't know how I did that; the other was due to a hardware problem causing my computer to shut down in the middle of an upgrade, but nevertheless it was possible to resume the update (in text mode) once the computer came back up). So I don't think updating Ubuntu is necessarily doomed to failure; although it doesn't seem to be quite as good as certain other distributions.
I thought the general Slashdot opinion (I have no idea whether it's right or not, I'm not a lawyer and I live in the wrong country to know American law) was that although downloading meant that copyright infringement had been committed, it was the person you downloaded from that made the copy, not you.
Over here in the UK, when something like that happens, we generally tell the media and let them and the corporation fight it out. (You sometimes even get your money back when that happens.) Complaining at large companies makes good TV over here, so the TV companies are generally happy to investigate; and although things don't get all that much better, at least they don't get worse.