Hmm, right, I was misled by what the guy said. I think that was the 486SX/DX instead... apart from that, what I said still applies. Just change 386 to 486...
It's not the same thing at all. What you describe is the process already currently in use in most chip factories: chips are categorized and sold according to some criteria; you talk about 386SX, where x386's would be sold as SX if the FP unit was tested defective, and DX otherwise. The same testing process has been in use for decades for a variety of chips. Some chips from the same batch are sold for different clock ratings, etc; memory chips can be sold with different capacities, and so on. But that's not the same thing: factories still sell chips they know for sure will work; they just make sure what to sell them as. Again, it's very different from selling chips that didn't pass QC as regular chips. It's a world of difference...
Interesting concept. Should consumers be lured into buying stuff that "should work, most of the time"? The fact that it's not "critical" (as in audio devices, etc) doesn't change anything. I think consumers already settle for too little.
There are some interesting research points around this (like error-recovery in multi-GHz logic chips, taking into account some temporary or permanent internal failure), but that's another idea entirely. But using defective chips? No way! Even if the failure is not apparent right away or undetectable - well, that means that we *don't know* what exactly will be wrong with that chip. All we know is that it has some fabrication defect - it's not in the state of working "the way we know it should work", and thus its failure rate becomes undecidable. Critical field or not, I want my chips to work. Period.
Anyway, if that ever happens, I think the Microsoft team will be very pleased. Unreliability as a core feature of some products - a dream come true. I'm kidding, but what I'm not kidding about is that it might create a "precedent" where it's ok to ship products that didn't pass the quality check. Someone joked about selling them to the 3rd world countries, but it's not really funny: it might very well happen. Why wouldn't it?
Ease of deployment is one of the biggest arguments in favor of Microsoft's products in the corporate world. But that's pretty much the deal with OSS solutions: whereas you have no license fees, you (usually) need more administration work time. A competent administrator can find ways of easing the deployment of Firefox in his particular environment. In the end, you have more security, more frequent updates and you keep the upper hand - something to consider in favor of OSS.
Hopefully not. Why not implant a device in every newborn that would call the police when someone *is about* to commit a crime, while we're at it? And then, while we're at it, why not kill them before they are born, if we can detect that they are bound to commit a crime sometime in their life? You guys are made to live in the "perfect" world depicted in "1984", aren't you?
Which is basically the exact same that is done for testing real drugs. In one case, you try to prove the drug has a benefit over the placebo. In the other case, you try to prove the placebo has no benefit over the drug that works. Which is a given, because that is how we test drugs. Uh huh.;-)
Nope, you're right. But that could mean they're not really up to the task of judging what is a flawed standard and what is not. It's not like they've never created a flawed de-facto standard themselves...
Yeah, exactly. This is getting old...
Besides, if what they claimed were really true, they wouldn't feel forced to say it all of the time. The superiority of their own products would shine. But since it doesn't work this way, because their products are clearly not superior, they choose to try explaining to the world that IT professionals are misled idiots blinded by the devil Linux. Actually, their whole way of handling the Linux "threat" is pretty insulting for whom it's aimed at - professionals, if you think about it. So what's next? When they are out of arguments, they will go on war against Linux and OSS? Maybe a real war, with soldiers and all. Yay.
As long as you don't use source code, yes you can use it as you please.
That is just not true! And especially not if you're a commercial vendor. Most OSS licenses forbid to distribute any software based on the open source projet in question without the source code to the whole, whether you have made modifications or not. Most vendors don't want to do that. Some OSS licenses would allow modifications without having to give them away (such as BSD-like licenses, I believe). Some forbid to even redistribute any binaries! Then, there are OSS licenses that forbid any commercial use of the software at all. And so on... So NO, you can't use it as you please. Not quite. Even though, as a random end-user, that's how it feels most of the time.
The license only covers source code.
Again, not true. The license covers the whole thing. The source code is an integral part of the "deal" with OSS, that doesn't mean only the source code is covered by the license.
Maybe you should read the whole recent thread about GPL violations to begin with... As for patents, no they are not essential to protect intellectual property. Copyrights are. Patents are other beasts with their own rationale. That would be interesting to start a new thread about patents... but then again, this is a recurring topic here, so we're going to have one any time soon...
Patents are not the answer to everything. Just because OSS stuff is not patented doesn't mean you can use it as you please. You're legally bound by a license. I guess if the Firefox team thought Microsoft was stealing core ideas, they would be entitled to sue. Even more so than OSS is basically the most public exposure you can get - which means, you have the more IP protection possible. So no, MS is *not* free to steal ideas. If there is a clear document from the Mozilla foundation lying around which talks about tabbed browsing, for instance, and well before MS shipped IE7, Mozilla could sue them. Contrary to a popular opinion, patents don't particularly protect intellectual property better. And no, MS *could not* patent tabbed browsing itself now, for instance (even if no one else has), because it was publicly exposed long ago in several software packages, both open and closed source.
Microsoft (and a lot of other companies) does use public research (in which we could categorize OSS, in a way) to make software, but they also give back by supporting research in a lot of areas. I'm not saying everything is all right, but let's get some perspective. If anything, MS could make donations to the Mozilla foundation (who knows if they don't already), and that would shut everyone's mouth...
You don't have to be a company to get a.fr domain name. It's just significantly more expensive to get a.fr domain than a.net or.com one... Besides, this woman probably is a "small company" since she works independent and as such is entitled to have claims on her name, be it her first name!
That said, it's likely that this domain name was chosen on purpose (I think her son took care of the web site) to get more audience. But then again, she's just doing very very small business. Sueing her was just ridiculous in my opinion. If the damn chocolate company wanted this domain name to begin with, they should have thought of registering it long ago instead of waiting for someone to register it and sue them. Lame, if you ask me.
I'm even pretty sure there are some GPL violators amongst the companies that fight against OSS (or at least trying to prove how bad and communist OSS is)...
There is: you have to be extra careful, all of the time. Secondly, you have to update everything on a very frequent basis (check at least once or twice a week). What it basically takes is: educated users and a decent administrator (if there are more than 5 to 10 computers in a LAN, I *strongly* suggest that there be someone in charge for administrating them).
In a work setting, most users *should* have restricted user rights on Windows (ie., no right to install software). And they should not read personal email on work computers. Period. With both these principles, pretty much nothing can go wrong... And most users don't need both these things to work. If software installation is needed - ask the administrator.
If you don't want to follow these, well, you just suffer. The company's choice, really.:)
Well, spyware will still get on a restricted account, but it won't be able to make as much damage. If it gets bad, just delete the kid's account and create a new one. You won't (usually) have to re-install Windows or spend hours cleaning it, which is a big relief.
"Sane person" and "average user joe" just doesn't add up. This was kind of a worst-case scenario, this was also meant to be funny - just in case you took it too seriously.
And by the way, of course no Windows user runs their desktop with Administrator rights? Uh huh! Why would it be any different if Linux became mainstream? It's not like it is any difficult to create users on Windows... yet practically nobody does that (I mean, maybe less than 5%?)
the ultimate step to Linux on the desktop would be some nice and warm adware - something that would pop up dozens of Windows at random times, all from nasty apps compiled against the commercial version of QT.:D Oh, and of course, it would break through your Linux firewall and recompile a modified Linux kernel behind your back. Does that sound nice already?
By the time you'll want to change processors, the motherboard will be obsolete anyway. Plus, it has good chances to have middle to poor performance compared to a AMD or Intel-only motherboard. I just don't see the point... To stay ahead of the competition, you have to provide high-performance products, not products that make people talk a lot but that no one will actually care about.
You learn that basically, you will spend lots of time and money just to drop the noise level a few dB's and with your system running at higher temperatures.
As for me, I've changed my stock Intel CPU fan for a Zalman 7700 Al-Cu and changed the case fans for Papst ones. The result was definitely encouraging, with the whole thing being less annoying to the ears that before. But it is nowhere near silent, still. So, I think this article's title is misleading to say the least. There is no way you can build a completely silent PC for now. Not with off-the-shelf parts anyway (you could always stick a giant passive heatsink to your CPU, with a part outside of the case, as with power amplifiers, and so on, but that's another story, and that would make your PC really huge and heavy).
I see your point, and you may be quite right. Weirdly enough, Firefox has "taken off" as a web browser in spite of that. If anything, for the average user, Mozilla Firefox was even less seen as a stable thing than the full Mozilla suite itself (which has been around for a long time now), and it was in fact suffering from the exact thing you're talking about, a forking that was not exactly easy to understand for everyone. Yet, a surprisingly high number of users are switching to it lately. This is something new here. But don't we get our hopes too high, I'm thinking it might be explained by the fact that more and more people actually don't care anymore what the stuff they use comes from. Maybe this is the beginning of a new "era": consumers starting not to care at all about brands and image anymore (in some way), and just considering the fact that it's cheap (or even free) and "works ok" for them. I mean, I don't believe one bit that the majority of users switching to Firefox did so because it was more compliant to standards, more secure, etc. I believe they did because it was free, and because there was this "new and exciting" bit about it. Then security comes second, in my opinion: but most people don't really care about security in itself; they are just happy to find a web browser that will not force them to reinstall Windows every month because it's gotten invaded with crap.
What does it have to do with closed source or not? If an open source project forks, it's usually for the project's sake. Major project decisions in a closed source setting can actually be made to maximize profit, to the expense of the project's quality. So, there is no definite answer to the question of stability in either context.
How legal exactly is it that Microsoft would forbid PC makers to sell their machines with any other OS if they plan on selling some of them with Windows pre-installed? Yet, this is exactly what has been the case. Is it fair competition? I don't think so! "You can't use our products if you use any other product." Oh yeah, what a wonderfully ethical practice...
Slashdot Mirror
Hmm, right, I was misled by what the guy said. I think that was the 486SX/DX instead... apart from that, what I said still applies. Just change 386 to 486...
It's not the same thing at all. What you describe is the process already currently in use in most chip factories: chips are categorized and sold according to some criteria; you talk about 386SX, where x386's would be sold as SX if the FP unit was tested defective, and DX otherwise. The same testing process has been in use for decades for a variety of chips. Some chips from the same batch are sold for different clock ratings, etc; memory chips can be sold with different capacities, and so on. But that's not the same thing: factories still sell chips they know for sure will work; they just make sure what to sell them as. Again, it's very different from selling chips that didn't pass QC as regular chips. It's a world of difference...
Interesting concept. Should consumers be lured into buying stuff that "should work, most of the time"? The fact that it's not "critical" (as in audio devices, etc) doesn't change anything. I think consumers already settle for too little.
There are some interesting research points around this (like error-recovery in multi-GHz logic chips, taking into account some temporary or permanent internal failure), but that's another idea entirely. But using defective chips? No way! Even if the failure is not apparent right away or undetectable - well, that means that we *don't know* what exactly will be wrong with that chip. All we know is that it has some fabrication defect - it's not in the state of working "the way we know it should work", and thus its failure rate becomes undecidable. Critical field or not, I want my chips to work. Period.
Anyway, if that ever happens, I think the Microsoft team will be very pleased. Unreliability as a core feature of some products - a dream come true. I'm kidding, but what I'm not kidding about is that it might create a "precedent" where it's ok to ship products that didn't pass the quality check. Someone joked about selling them to the 3rd world countries, but it's not really funny: it might very well happen. Why wouldn't it?
That's amazing all these people speaking weird languages huh?
Ease of deployment is one of the biggest arguments in favor of Microsoft's products in the corporate world. But that's pretty much the deal with OSS solutions: whereas you have no license fees, you (usually) need more administration work time. A competent administrator can find ways of easing the deployment of Firefox in his particular environment. In the end, you have more security, more frequent updates and you keep the upper hand - something to consider in favor of OSS.
Hopefully not. Why not implant a device in every newborn that would call the police when someone *is about* to commit a crime, while we're at it? And then, while we're at it, why not kill them before they are born, if we can detect that they are bound to commit a crime sometime in their life? You guys are made to live in the "perfect" world depicted in "1984", aren't you?
Which is basically the exact same that is done for testing real drugs. In one case, you try to prove the drug has a benefit over the placebo. In the other case, you try to prove the placebo has no benefit over the drug that works. Which is a given, because that is how we test drugs. Uh huh. ;-)
Nope, you're right. But that could mean they're not really up to the task of judging what is a flawed standard and what is not. It's not like they've never created a flawed de-facto standard themselves...
And apart from the processor, there's nothing really new in this thing.
It's about time this half-baked tool they call a language disappeared...
And one of the reasons I've switched to self-employed...
Yeah, exactly. This is getting old... Besides, if what they claimed were really true, they wouldn't feel forced to say it all of the time. The superiority of their own products would shine. But since it doesn't work this way, because their products are clearly not superior, they choose to try explaining to the world that IT professionals are misled idiots blinded by the devil Linux. Actually, their whole way of handling the Linux "threat" is pretty insulting for whom it's aimed at - professionals, if you think about it. So what's next? When they are out of arguments, they will go on war against Linux and OSS? Maybe a real war, with soldiers and all. Yay.
That is just not true! And especially not if you're a commercial vendor. Most OSS licenses forbid to distribute any software based on the open source projet in question without the source code to the whole, whether you have made modifications or not. Most vendors don't want to do that. Some OSS licenses would allow modifications without having to give them away (such as BSD-like licenses, I believe). Some forbid to even redistribute any binaries! Then, there are OSS licenses that forbid any commercial use of the software at all. And so on... So NO, you can't use it as you please. Not quite. Even though, as a random end-user, that's how it feels most of the time.
The license only covers source code.Again, not true. The license covers the whole thing. The source code is an integral part of the "deal" with OSS, that doesn't mean only the source code is covered by the license.
Maybe you should read the whole recent thread about GPL violations to begin with... As for patents, no they are not essential to protect intellectual property. Copyrights are. Patents are other beasts with their own rationale. That would be interesting to start a new thread about patents... but then again, this is a recurring topic here, so we're going to have one any time soon...
Patents are not the answer to everything. Just because OSS stuff is not patented doesn't mean you can use it as you please. You're legally bound by a license. I guess if the Firefox team thought Microsoft was stealing core ideas, they would be entitled to sue. Even more so than OSS is basically the most public exposure you can get - which means, you have the more IP protection possible. So no, MS is *not* free to steal ideas. If there is a clear document from the Mozilla foundation lying around which talks about tabbed browsing, for instance, and well before MS shipped IE7, Mozilla could sue them. Contrary to a popular opinion, patents don't particularly protect intellectual property better. And no, MS *could not* patent tabbed browsing itself now, for instance (even if no one else has), because it was publicly exposed long ago in several software packages, both open and closed source.
Microsoft (and a lot of other companies) does use public research (in which we could categorize OSS, in a way) to make software, but they also give back by supporting research in a lot of areas. I'm not saying everything is all right, but let's get some perspective. If anything, MS could make donations to the Mozilla foundation (who knows if they don't already), and that would shut everyone's mouth...
You don't have to be a company to get a .fr domain name. It's just significantly more expensive to get a .fr domain than a .net or .com one... Besides, this woman probably is a "small company" since she works independent and as such is entitled to have claims on her name, be it her first name!
That said, it's likely that this domain name was chosen on purpose (I think her son took care of the web site) to get more audience. But then again, she's just doing very very small business. Sueing her was just ridiculous in my opinion. If the damn chocolate company wanted this domain name to begin with, they should have thought of registering it long ago instead of waiting for someone to register it and sue them. Lame, if you ask me.
I'm even pretty sure there are some GPL violators amongst the companies that fight against OSS (or at least trying to prove how bad and communist OSS is)...
"Honesty is such a lonely word..."
There is: you have to be extra careful, all of the time. Secondly, you have to update everything on a very frequent basis (check at least once or twice a week). What it basically takes is: educated users and a decent administrator (if there are more than 5 to 10 computers in a LAN, I *strongly* suggest that there be someone in charge for administrating them).
In a work setting, most users *should* have restricted user rights on Windows (ie., no right to install software). And they should not read personal email on work computers. Period. With both these principles, pretty much nothing can go wrong... And most users don't need both these things to work. If software installation is needed - ask the administrator.
If you don't want to follow these, well, you just suffer. The company's choice, really. :)
Well, spyware will still get on a restricted account, but it won't be able to make as much damage. If it gets bad, just delete the kid's account and create a new one. You won't (usually) have to re-install Windows or spend hours cleaning it, which is a big relief.
"Sane person" and "average user joe" just doesn't add up. This was kind of a worst-case scenario, this was also meant to be funny - just in case you took it too seriously.
And by the way, of course no Windows user runs their desktop with Administrator rights? Uh huh! Why would it be any different if Linux became mainstream? It's not like it is any difficult to create users on Windows... yet practically nobody does that (I mean, maybe less than 5%?)
the ultimate step to Linux on the desktop would be some nice and warm adware - something that would pop up dozens of Windows at random times, all from nasty apps compiled against the commercial version of QT. :D Oh, and of course, it would break through your Linux firewall and recompile a modified Linux kernel behind your back. Does that sound nice already?
By the time you'll want to change processors, the motherboard will be obsolete anyway. Plus, it has good chances to have middle to poor performance compared to a AMD or Intel-only motherboard. I just don't see the point... To stay ahead of the competition, you have to provide high-performance products, not products that make people talk a lot but that no one will actually care about.
You learn that basically, you will spend lots of time and money just to drop the noise level a few dB's and with your system running at higher temperatures.
As for me, I've changed my stock Intel CPU fan for a Zalman 7700 Al-Cu and changed the case fans for Papst ones. The result was definitely encouraging, with the whole thing being less annoying to the ears that before. But it is nowhere near silent, still. So, I think this article's title is misleading to say the least. There is no way you can build a completely silent PC for now. Not with off-the-shelf parts anyway (you could always stick a giant passive heatsink to your CPU, with a part outside of the case, as with power amplifiers, and so on, but that's another story, and that would make your PC really huge and heavy).
I see your point, and you may be quite right. Weirdly enough, Firefox has "taken off" as a web browser in spite of that. If anything, for the average user, Mozilla Firefox was even less seen as a stable thing than the full Mozilla suite itself (which has been around for a long time now), and it was in fact suffering from the exact thing you're talking about, a forking that was not exactly easy to understand for everyone. Yet, a surprisingly high number of users are switching to it lately. This is something new here. But don't we get our hopes too high, I'm thinking it might be explained by the fact that more and more people actually don't care anymore what the stuff they use comes from. Maybe this is the beginning of a new "era": consumers starting not to care at all about brands and image anymore (in some way), and just considering the fact that it's cheap (or even free) and "works ok" for them. I mean, I don't believe one bit that the majority of users switching to Firefox did so because it was more compliant to standards, more secure, etc. I believe they did because it was free, and because there was this "new and exciting" bit about it. Then security comes second, in my opinion: but most people don't really care about security in itself; they are just happy to find a web browser that will not force them to reinstall Windows every month because it's gotten invaded with crap.
What does it have to do with closed source or not? If an open source project forks, it's usually for the project's sake. Major project decisions in a closed source setting can actually be made to maximize profit, to the expense of the project's quality. So, there is no definite answer to the question of stability in either context.
How legal exactly is it that Microsoft would forbid PC makers to sell their machines with any other OS if they plan on selling some of them with Windows pre-installed? Yet, this is exactly what has been the case. Is it fair competition? I don't think so! "You can't use our products if you use any other product." Oh yeah, what a wonderfully ethical practice...