But there are some silly things in here, too. First, I gotta ask about these two "broken" RH distros...like, uh, broken how? Like, "the CD was broken in the box" or like "couldn't get through the installer" or like "wouldn't boot on any computer I tried it on" or what? Or just "couldn't get a productive desktop system easily" or "couldn't get sendmail, apache, mysql, etc. on there and happy easily enough for my taste"?
I don't think I can remember any distro I've tried that I would have said was "broken" really...they all did something....that's just weird.
And these comments about professional windows admins who no better, seriously...get real. I mean, I'm a *nix zealot, no fooling (check other posts). But it's not like Windows is completely incapable of doing lots of jobs that professional admins require...quite the opposite. There are viable (if not maybe as secure) servers for everything, and if the boss is buying...whatever. A lot of admins that are tremendously put upon and don't have office clout can't sit there and refuse that as "bad concrete", offering up some "free" alternative. What do they do, just get another job?
Some of us can take the high road in one form or another if we choose, but not without some cost.
And lots of us probably sit around in jobs that force us to work with windows, waiting and looking for the job that won't.
But generally speaking, acting the way you prescribe is not healthy from an employment perspective.
Not at all the point. This is neat for a certain purpose. That purpose has nothing to do with a laptop.
It would be nice for a machine you have no physical access to, like, say, a virtual co-lo gig that preinstalls redhat for cheaper than bsd. If you have a laptop, you will have physical access to the machine, so doing an elaborate procedure like this from another computer is just downright silly.
You argue against this person (The italisized portion of your post) who shows a *parallel* situation, by demonstrating the way an *opposite* situation behaves!
Seriously, he says "if it was about 'upgrading' windows boxes to Linux it would not be considered flamebait." THIS IS NOT THE OPPOSITE OF THIS ARTICLE! It is in fact a parallel; the same thing with the names changed.
Now, as it happens, this act (Linux "upgraded" to FreeBSD) has no real opposite. Even a FreeBSD->Linux upgrade would still be more the same than opposite. But that just makes your argument all the more silly.
Now, that's not to say I agree with this fool, either. Frankly, you're both wrong; you for your argument, and he for his conclusion. Any of the three parallels would be flamebait...no matter what, an article written up this provocatively and on this subject is going to cause the vast majority of responses to be holy-war. You don't think an artical about "upgrading" Windows to Linux would be flame-laden? Who the hell are you kidding?
On the other hand, that's why I clicked on the story; I wanted to see it happen, and work my logic muscle a bit. Against both sides. I mean, seriously...you get what you pay for. Who here didn't know they were about to read some flames?
While I certainly get your main point (rooting one box will leave the rest safe) I simply *must* take issue with your example.
You say if this guy roots your DNS VM, he won't be able to deface your website. I'll point out the obvious: he now has control over the web address, and can point your website at his own box, where the defaced site lies. Or he can point it at the DNS box itself, install apache, and deface it there.
Point is, if he roots your DNS server, you are all kinds of jacked.
"But Windows can seemlessly change between SCSI-emulation and IDE, without requiring boottime option (allthough a reboot is required for installation)."
This sounds very unlikely to me. I admit, I don't really know, having not owned a windows computer in a few years...but I can't see any conceivable way this could be true. Does windows have some right-click option on the drive letter that has a check-box for "use scsi-emulation" or something?
I think it is much much more likely that either a)windows leaves the drive in scsi-emu mode all the time, or b)windows loads normal ide stuff, and nero/roxio/whatever loads up the scsi-emu.
The big question is "can you really tell windows to turn scsi-emu on/off?" I doubt it.
"If we live is a so-called capitalistic or free market society why do we even have rules like this?"
Well, first, nobody lives in a "perfectly" capitalist society. All systems currently in place are some blend of capitalism and socialism. Thank God.
But second, the reason rules like this are in place is that it encourages people to *share* ideas. In a world without patents, breakthrough ideas would be jeallously gaurded, because others would steal the idea and use it themselves. Patents allow inventors to publicize their ideas, safe in the knowledge that they have a (limited-time) legal protection. And the patent document forces them to completely explain it, so that other people can use the idea to create newer, better ideas, which they can then patent. This is all set out in the U.S. Constitution, but that was not the first instance of such things.
Now, as for whether *software* should be patentable.....that's a lot harder question. I like to say it shouldn't, because ultimately these patents aren't protecting the idea (implementation) so much as the ability to perform some task. In other words, if you had a patent on a bicycle, I could build a bike using something other than a chain drive system and be clear. But if you obtain patent on a "shopping-cart" app, my cart may be implemented completely differently, but since it *does* the same thing, I can't use it.
In other words, this is like a patent filed by the guy who invented the bicycle, but which is said to cover all "two-wheeled people movers."
Yeah...that's interesting. You know they always say how the shuttle travels at like mach 30 or whatever...but seeing as it's out in space, where sound doesn't travel...wouldn't it be "mach infinity"?
"I'm sorry, but could you point to the part of copyright or case law that says that?"
No, because it isn't a matter of case law; it's WRITTEN EXPLICITLY IN THE GPL.
And if you read any of Linus or Stallman's comments, they absolutely do not ever suggest that anything must be GPL that doesn't either incorporate GPL code or statically link to GPL code.
Whatever yourself. First off, there isn't significant evidence of that ever happening, just wild speculations from MS execs. The only place I know of (Thailand) that sold a significant number of Linux desktops and then studied their retention found that the vast majority kept Linux, even though street prices for pirate Windows were ~$5. They also sell Windows versions, at nearly the same price (MS gave them XP+OfficeXP for ~$35/seat) but those are not selling nearly as well.
Secondly, there is no way in hell that the number of Linux-sold PC's getting Windows put on them will equal the number of Windows-sold PC's getting Linux any time soon.
"Yes, I have. Lots of others have done this too. Being able to look up the source in case of network troubles is really a great help. The debug mode of the driver is very helpful, the source even better."
Huh, huh. Yeah, ok man. I know lots of people who are capable of doing what you describe, and I know no one who actually has. Everyone I've known to look at driver code has done it for a)fun or b)active development of experimental drivers.
I can see looking into the code if you are having trouble with a driver that is brand new or known-buggy. But if you are looking at the tulip.o driver to help figure out your network problem, you have a screw loose. I admit that I have never read driver source. I've run Linux for over 5 years and solved a multitude of problems ranging from the simple to obscure. It never required looking at the code.
And thank heavens...it would be a really shitty system that required you to read its source code to solve normal problems.
Sorry for the shouting, but I'm starting to get annoyed as I read this over and over in this story. There is ABSOLUTELY NO AMBIGUITY in the GPL concerning what code must be re-GPLed and what avoids that restriction. Anything into which you copied previously GPLed code (that wasn't also available under a different license), or which you statically-linked to GPL code, MUST BE GPL.
If you didn't do either of those two things, you are safe!
"it's hard to tell where their drivers start and the kernel begins"
Um, yeah, I can see how that would be a problem. Here, let me help:
The drivers start at the beginning of the files you downloaded from Nvidia, and end at the end of those files. The kernel pretty much includes everything that was under/usr/src/linux before you installed the Nvidia drivers.
After the installation, the nvidia modules are two files...a binary kernel module, and a binary XF86 module. The kernel is generally a single image at this point.
I can see how this could be tricky...
So anyway, in the larger sense, I agree with you that binaries ought to be allowed. However, I disagree that there is any question where the lines are drawn; if you use GPL code in your stuff, or statically link to it, your code must be GPL. If it doesn't, it doesn't have to be. That's really not that complicated, assuming you paid attention when you wrote the thing.
"You and 7658880 are both right in saying nVidia hasn't given anything (of value) to Linux."
No, they are both wrong. Nvidia has given an excellent and fully-usable line of video chipsets to Linux.
That's the big problem here; the module isn't the product! The hardware is...and Nvidia's cards are excellent for running linux. Millions of people (and presumably a large percentage of this year's first-time linux users) already own them. So the fact that they made those boards useable means Linux can reach a much wider audience than it could before, and thus Nvidia *did* give something to Linux.
Well, I appreciate the information concerning the amounts of OSS/non-OSS code in those drivers.
However, I think the fact that you can't compile these for a SPARC is a minor problem at worst; don't most of those boxes come with their own specialized video systems? I have to guess they don't come with nvidia gear (unless by some serious perversity, nvidia makes a driver for Solaris, which seems highly unlikely) which means there isn't much use for the drivers there. In contrast, millions of nvidia boards are shipped in the x86 world annually, and this driverset runs on pretty much any x86 linux system. Or at least the ones I've tried.
And anyway, I don't see how that's relevant. This company owns these drivers. They wrote them. They allege that there is IP in the binary part that they aren't willing to let competitors see, but they could just as easily say "we don't feel like it." It's their stuff. So we have a really simple choice...use them or don't. Now, since those drivers have in my experience been as stable and easy to install as any other video drivers I have ever used in Linux, and since my goal is actually using the system, rather than holding idealogical bullshit arguments on/., I use them.
It is worth noting that I have had about 5 times as much trouble getting ATI gear to work right.
Oh yeah, and thanks a lot for the baseless insult about sounding like a Windows user. For the record, the last time I had Windows on one of my own computers was in 1999. Now, I still don't run around senselessly bashing those who do use it...
Right, and there's another semester this year. So, there's about a whole semester left for this jackass to continue needlessly and senselessly abusing his parents.
It might make sense to take that position, if such a thing as a "module vendor" existed. As it happens, it doesn't, and no one is out trying to sell binary modules for Linux. The creators of binary modules are *hardware vendors* and they are "contributing" by making their hardware compatible with the free system.
This is not parasitic; if they want, they can just not bother, and you can just not use that hardware in Linux. Let's not forget, it's not like you wrote the driver; why would you want to keep people from making their hardware usable on your system? If a manufacturer says "well, sorry, I want to support linux, but not if it means letting the competition get a sneak peak at this crazy technology in my drivers" you would just say, "ok, parasite, we don't need your stupid hardware."
When the manufacturer in question is a leading producer of video boards, such fanaticism is extremely foolish.
Well, the trouble is, the drivers to products that really are trivial (those NICs you mentioned) are already available, because those companies agreed with you and released drivers openly or at least released information, allowing the community to produce them.
But some products aren't that way; nvidia, for instance, at least *says* they have IP tied up in the binary part of their drivers that they can't afford to let competitors (ATI) get ahold of. I don't know whether it's the truth.....I haven't seen the code!
Slashdot Mirror
Dude, don't hold back ;-)
Your main theme I agree with:
(BSD != Dying) && (Linux == OK_AS_WELL) && (Windows == BUSTICATED)
But there are some silly things in here, too. First, I gotta ask about these two "broken" RH distros...like, uh, broken how? Like, "the CD was broken in the box" or like "couldn't get through the installer" or like "wouldn't boot on any computer I tried it on" or what? Or just "couldn't get a productive desktop system easily" or "couldn't get sendmail, apache, mysql, etc. on there and happy easily enough for my taste"?
I don't think I can remember any distro I've tried that I would have said was "broken" really...they all did something....that's just weird.
And these comments about professional windows admins who no better, seriously...get real. I mean, I'm a *nix zealot, no fooling (check other posts). But it's not like Windows is completely incapable of doing lots of jobs that professional admins require...quite the opposite. There are viable (if not maybe as secure) servers for everything, and if the boss is buying...whatever. A lot of admins that are tremendously put upon and don't have office clout can't sit there and refuse that as "bad concrete", offering up some "free" alternative. What do they do, just get another job?
Some of us can take the high road in one form or another if we choose, but not without some cost.
And lots of us probably sit around in jobs that force us to work with windows, waiting and looking for the job that won't.
But generally speaking, acting the way you prescribe is not healthy from an employment perspective.
---
Not at all the point. This is neat for a certain purpose. That purpose has nothing to do with a laptop.
It would be nice for a machine you have no physical access to, like, say, a virtual co-lo gig that preinstalls redhat for cheaper than bsd. If you have a laptop, you will have physical access to the machine, so doing an elaborate procedure like this from another computer is just downright silly.
If only I had a modpoint.....goddamn that's funny.
Ridiculous.
You argue against this person (The italisized portion of your post) who shows a *parallel* situation, by demonstrating the way an *opposite* situation behaves!
Seriously, he says "if it was about 'upgrading' windows boxes to Linux it would not be considered flamebait." THIS IS NOT THE OPPOSITE OF THIS ARTICLE! It is in fact a parallel; the same thing with the names changed.
Now, as it happens, this act (Linux "upgraded" to FreeBSD) has no real opposite. Even a FreeBSD->Linux upgrade would still be more the same than opposite. But that just makes your argument all the more silly.
Now, that's not to say I agree with this fool, either. Frankly, you're both wrong; you for your argument, and he for his conclusion. Any of the three parallels would be flamebait...no matter what, an article written up this provocatively and on this subject is going to cause the vast majority of responses to be holy-war. You don't think an artical about "upgrading" Windows to Linux would be flame-laden? Who the hell are you kidding?
On the other hand, that's why I clicked on the story; I wanted to see it happen, and work my logic muscle a bit. Against both sides. I mean, seriously...you get what you pay for. Who here didn't know they were about to read some flames?
While I certainly get your main point (rooting one box will leave the rest safe) I simply *must* take issue with your example.
You say if this guy roots your DNS VM, he won't be able to deface your website. I'll point out the obvious: he now has control over the web address, and can point your website at his own box, where the defaced site lies. Or he can point it at the DNS box itself, install apache, and deface it there.
Point is, if he roots your DNS server, you are all kinds of jacked.
"But Windows can seemlessly change between SCSI-emulation and IDE, without requiring boottime option (allthough a reboot is required for installation)."
This sounds very unlikely to me. I admit, I don't really know, having not owned a windows computer in a few years...but I can't see any conceivable way this could be true. Does windows have some right-click option on the drive letter that has a check-box for "use scsi-emulation" or something?
I think it is much much more likely that either a)windows leaves the drive in scsi-emu mode all the time, or b)windows loads normal ide stuff, and nero/roxio/whatever loads up the scsi-emu.
The big question is "can you really tell windows to turn scsi-emu on/off?" I doubt it.
"If we live is a so-called capitalistic or free market society why do we even have rules like this?"
Well, first, nobody lives in a "perfectly" capitalist society. All systems currently in place are some blend of capitalism and socialism. Thank God.
But second, the reason rules like this are in place is that it encourages people to *share* ideas. In a world without patents, breakthrough ideas would be jeallously gaurded, because others would steal the idea and use it themselves. Patents allow inventors to publicize their ideas, safe in the knowledge that they have a (limited-time) legal protection. And the patent document forces them to completely explain it, so that other people can use the idea to create newer, better ideas, which they can then patent. This is all set out in the U.S. Constitution, but that was not the first instance of such things.
Now, as for whether *software* should be patentable.....that's a lot harder question. I like to say it shouldn't, because ultimately these patents aren't protecting the idea (implementation) so much as the ability to perform some task. In other words, if you had a patent on a bicycle, I could build a bike using something other than a chain drive system and be clear. But if you obtain patent on a "shopping-cart" app, my cart may be implemented completely differently, but since it *does* the same thing, I can't use it.
In other words, this is like a patent filed by the guy who invented the bicycle, but which is said to cover all "two-wheeled people movers."
Yeah...that's interesting. You know they always say how the shuttle travels at like mach 30 or whatever...but seeing as it's out in space, where sound doesn't travel...wouldn't it be "mach infinity"?
Yar.
"...for example an overnight flight from East Asia to Europe would arrive too early in the morning, before the airports open..."
er...airports close?
Wow.
Reading through, I see what you mean. Thanks a lot for that post. Next time I get some modpoints...uh...
"What about dynamically linking to code..."
You just answered your own question. Is it statically linked? Does it incorporate GPL code?
No, and no, so I'm pretty sure you're fine.
"I'm sorry, but could you point to the part of copyright or case law that says that?"
No, because it isn't a matter of case law; it's WRITTEN EXPLICITLY IN THE GPL.
And if you read any of Linus or Stallman's comments, they absolutely do not ever suggest that anything must be GPL that doesn't either incorporate GPL code or statically link to GPL code.
Whatever yourself. First off, there isn't significant evidence of that ever happening, just wild speculations from MS execs. The only place I know of (Thailand) that sold a significant number of Linux desktops and then studied their retention found that the vast majority kept Linux, even though street prices for pirate Windows were ~$5. They also sell Windows versions, at nearly the same price (MS gave them XP+OfficeXP for ~$35/seat) but those are not selling nearly as well.
Secondly, there is no way in hell that the number of Linux-sold PC's getting Windows put on them will equal the number of Windows-sold PC's getting Linux any time soon.
"Yes, it is parasitic,..."
Er, were you going to support that at all? I said why it wasn't...
yes...biding their time with their video technology. "We'll release this in a few years!"
;P
He must be talking about bitboys
"when forced to chose between shitty drivers and binaries drivers, I'll choose binaries, thank you. I've had my fill of nv."
HAHAHAHAHA!!!! Yeah. Damn straight.
"Yes, I have. Lots of others have done this too. Being able to look up the source in case of network troubles is really a great help. The debug mode of the driver is very helpful, the source even better."
Huh, huh. Yeah, ok man. I know lots of people who are capable of doing what you describe, and I know no one who actually has. Everyone I've known to look at driver code has done it for a)fun or b)active development of experimental drivers.
I can see looking into the code if you are having trouble with a driver that is brand new or known-buggy. But if you are looking at the tulip.o driver to help figure out your network problem, you have a screw loose. I admit that I have never read driver source. I've run Linux for over 5 years and solved a multitude of problems ranging from the simple to obscure. It never required looking at the code.
And thank heavens...it would be a really shitty system that required you to read its source code to solve normal problems.
Aren't the headers all LGPL?
BUT THAT ISN'T A PROBLEM AT ALL!!!!
Sorry for the shouting, but I'm starting to get annoyed as I read this over and over in this story. There is ABSOLUTELY NO AMBIGUITY in the GPL concerning what code must be re-GPLed and what avoids that restriction. Anything into which you copied previously GPLed code (that wasn't also available under a different license), or which you statically-linked to GPL code, MUST BE GPL.
If you didn't do either of those two things, you are safe!
This is not that fucking complicated.
"it's hard to tell where their drivers start and the kernel begins"
/usr/src/linux before you installed the Nvidia drivers.
Um, yeah, I can see how that would be a problem. Here, let me help:
The drivers start at the beginning of the files you downloaded from Nvidia, and end at the end of those files. The kernel pretty much includes everything that was under
After the installation, the nvidia modules are two files...a binary kernel module, and a binary XF86 module. The kernel is generally a single image at this point.
I can see how this could be tricky...
So anyway, in the larger sense, I agree with you that binaries ought to be allowed. However, I disagree that there is any question where the lines are drawn; if you use GPL code in your stuff, or statically link to it, your code must be GPL. If it doesn't, it doesn't have to be. That's really not that complicated, assuming you paid attention when you wrote the thing.
"You and 7658880 are both right in saying nVidia hasn't given anything (of value) to Linux."
No, they are both wrong. Nvidia has given an excellent and fully-usable line of video chipsets to Linux.
That's the big problem here; the module isn't the product! The hardware is...and Nvidia's cards are excellent for running linux. Millions of people (and presumably a large percentage of this year's first-time linux users) already own them. So the fact that they made those boards useable means Linux can reach a much wider audience than it could before, and thus Nvidia *did* give something to Linux.
Well, I appreciate the information concerning the amounts of OSS/non-OSS code in those drivers.
/., I use them.
However, I think the fact that you can't compile these for a SPARC is a minor problem at worst; don't most of those boxes come with their own specialized video systems? I have to guess they don't come with nvidia gear (unless by some serious perversity, nvidia makes a driver for Solaris, which seems highly unlikely) which means there isn't much use for the drivers there. In contrast, millions of nvidia boards are shipped in the x86 world annually, and this driverset runs on pretty much any x86 linux system. Or at least the ones I've tried.
And anyway, I don't see how that's relevant. This company owns these drivers. They wrote them. They allege that there is IP in the binary part that they aren't willing to let competitors see, but they could just as easily say "we don't feel like it." It's their stuff. So we have a really simple choice...use them or don't. Now, since those drivers have in my experience been as stable and easy to install as any other video drivers I have ever used in Linux, and since my goal is actually using the system, rather than holding idealogical bullshit arguments on
It is worth noting that I have had about 5 times as much trouble getting ATI gear to work right.
Oh yeah, and thanks a lot for the baseless insult about sounding like a Windows user. For the record, the last time I had Windows on one of my own computers was in 1999. Now, I still don't run around senselessly bashing those who do use it...
Right, and there's another semester this year. So, there's about a whole semester left for this jackass to continue needlessly and senselessly abusing his parents.
Er, yeah, that's a little warped.
It might make sense to take that position, if such a thing as a "module vendor" existed. As it happens, it doesn't, and no one is out trying to sell binary modules for Linux. The creators of binary modules are *hardware vendors* and they are "contributing" by making their hardware compatible with the free system.
This is not parasitic; if they want, they can just not bother, and you can just not use that hardware in Linux. Let's not forget, it's not like you wrote the driver; why would you want to keep people from making their hardware usable on your system? If a manufacturer says "well, sorry, I want to support linux, but not if it means letting the competition get a sneak peak at this crazy technology in my drivers" you would just say, "ok, parasite, we don't need your stupid hardware."
When the manufacturer in question is a leading producer of video boards, such fanaticism is extremely foolish.
Well, the trouble is, the drivers to products that really are trivial (those NICs you mentioned) are already available, because those companies agreed with you and released drivers openly or at least released information, allowing the community to produce them.
But some products aren't that way; nvidia, for instance, at least *says* they have IP tied up in the binary part of their drivers that they can't afford to let competitors (ATI) get ahold of. I don't know whether it's the truth.....I haven't seen the code!