It does require your contribution to be acknowledged, IIRC. Thus, if someone wants to know how Person X accomplished Task Y, and they refer to Person Z as having contributed code for Y under the BSD license, then the curious can look to Person Z for details.
i.e. If you write code, and someone else uses it, you're still acknowledged for it and people can look right to the source. Someone using your code in a non-free context does not prevent you from using it in a free context.
If I'm not mistaken, one of the main reasons the BSD license gets slammed by dedicated GPL proponents is because the use of BSD-licensed code does not demand that the resulting code also be BSD-licensed. Thus, if MS made any changes to the code(i.e. making it work with Windows), they can claim it as theirs and use their own licensing scheme - something which the GPL would not allow, but which the BSD license, if I understand the relevant points correctly, does.
So, technically, distributing ftp.exe would be against Microsoft's licensing and would be a breach of copyright, unless they have specifically stated that ftp.exe is under a different license. It would indeed be illegal. It's not something they're likely to pursue, though(not least because, with the number of free FTP clients both graphical and text out there, it's not likely to become a problem).
This has probably been mentioned already, but: the distinction arises in distribution. You can make a copy for yourself; fine. But if you sell your copy, if you make multiple copies to give or sell, if you play your recording at a public event, then you are in the wrong.
I think that's the gist of it, anyway; someone with better access to the details please feel free to correct or confirm.
A reasonable attempt should be made at keeping dangerous goods from being applied in a dangerous manner. Drugs, guns... However, as has already been stated, there are a number of organisations in the Western world that specifically want the ability to censor their Internet connection, for instance. Schools might do it, parents might do it, whether for similar, general reasons(e.g. to keep children from viewing adult material) or something more specific(parents don't want children going to a particular website). Corporations do it to try to prevent their network from being used for non-work purposes - and as they pay for the maintenance, bandwidth, electricity, and so forth, that's their prerogative. If the same restrictions let a totalitarian regime do nasty things to their people, yes, it's nasty, but it's rather hard for the software developers to prevent it while satisfying the people who want those controls closer to home.
Anything can be exploited. A chair can do a fair bit of damage to someone, but they aren't outlawed because of it. Something that makes it very easy to do harm should be controlled(a driver's license for vehicles; I personally think there should be a permit to carry a gun if there isn't already), but you can't keep people from using something good to do something bad - and some people do see the ability to e.g. censor the Internet as "something good", even though it does involve Windows.
"Guns don't kill people, people kill people" has a grain of truth to it - someone has to actually pick up that gun and fire it - but is all too often taken to excess. No, the firearm company should not be blamed - but the people distributing those guns ought to have a care who they sell guns to. Someone who mutters ominously about other guys looking at his wife while he's browsing the selection might need a small reminder about the law...
As I've said elsewhere, the official nvidia drivers caused my system to freeze and eventually reboot when I tried to load X. Nothing in the logs, they just stopped.
Most likely because it's rather hard to prove that you collected that data on your own. Logically proving a negative is extremely difficult if not impossible.
top worked for me, but I found that any other tool did not, such as gtop or the KDE usage-monitor applet included. I didn't actually get the new version installed last night(I'm going to get a network install running before I head to work) so I don't know if this has changed.
IIRC, it's noted in the handbook that if the kernel and the source tree are different versions, then such utilities as top may not properly function. It could be that these other ports haven't been re-ported to the same kernel version, and thus don't know how to get what they want from the kernel(or, depending, they may be ported to a newer kernel than the one you have installed, with the same result). I am very much a novice at programming, nowhere near the skill required to poke at the core code of my OS, so I don't know a way to verify this, much less fix it.
Hmmm, NVIDIA's driver didn't like my system, either - to the point that when X tried to start with it, my system would totally lock up. The open-source driver works much better for me.
And the logs didn't show any sort of error - everything just kinda stopped - so there wasn't much for me to submit for a bug report. Ah well, if it happens again I'll just try to be less apathetic.
My computer is a semi-old and quite unreliable pile o' junk. That Windows hates it is no surprise(though Windows actually does support my nVidia GeForce card's 3D acceleration). I've tried Red Hat, two versions of Mandrake, and Debian.
FreeBSD is the only thing I've tried that'd keep running if I didn't poke at it. And when I did choose to poke at it, it was most tolerant of it, and - thanks in large part to the devfs system - it's FAR easier to tell what I should be poking AT. Especially for my USB card reader - attach the device, and there it is, a brand new entry,/dev/ums0. Much easier than muddling through a whole tangle of device nodes and hoping that one of them is what I'm looking for.
And for all the extra time it takes, I'm very fond of the ports tree's default-ish approach of "compile from source to suit the system". My Linux experience was fraught with library conflicts in binary packages; in FreeBSD I've hit a few snags, but they were much more easily resolved - although the process was time-consuming, it was not terribly attention-consuming.
For a supposedly dead OS, FreeBSD lives quite well indeed on my system, when the Linux distros I've tried all died in short order. If only I had the space to compile OpenOffice, I'd be set.
Now I just hope the review hasn't been/.'d by the time I get home from work(.com is blocked by the firewall,.org is not. Maybe there's a/.er on my IT staff?).
Even machines, though, often follow this strategy. Most dishwashers I'm aware of have a two-stage wash at least(one bin of soap being added when the door is first closed, the other some time later), and the larger washers at laundromats soak the clothes for a while before they ask for any soap to be added at all.
However, I don't know if people would be terribly impressed by a computer that, when instructed to compile, instead started trying to force-crack passwords to make sure they're secure, or some other unrelated task. Still... "make -j4" If it can't compile ONE file right away, why not work on another?
Slashdot Mirror
It does require your contribution to be acknowledged, IIRC. Thus, if someone wants to know how Person X accomplished Task Y, and they refer to Person Z as having contributed code for Y under the BSD license, then the curious can look to Person Z for details.
i.e. If you write code, and someone else uses it, you're still acknowledged for it and people can look right to the source. Someone using your code in a non-free context does not prevent you from using it in a free context.
If I'm not mistaken, one of the main reasons the BSD license gets slammed by dedicated GPL proponents is because the use of BSD-licensed code does not demand that the resulting code also be BSD-licensed. Thus, if MS made any changes to the code(i.e. making it work with Windows), they can claim it as theirs and use their own licensing scheme - something which the GPL would not allow, but which the BSD license, if I understand the relevant points correctly, does.
So, technically, distributing ftp.exe would be against Microsoft's licensing and would be a breach of copyright, unless they have specifically stated that ftp.exe is under a different license. It would indeed be illegal. It's not something they're likely to pursue, though(not least because, with the number of free FTP clients both graphical and text out there, it's not likely to become a problem).
This has probably been mentioned already, but: the distinction arises in distribution. You can make a copy for yourself; fine. But if you sell your copy, if you make multiple copies to give or sell, if you play your recording at a public event, then you are in the wrong.
I think that's the gist of it, anyway; someone with better access to the details please feel free to correct or confirm.
A reasonable attempt should be made at keeping dangerous goods from being applied in a dangerous manner. Drugs, guns... However, as has already been stated, there are a number of organisations in the Western world that specifically want the ability to censor their Internet connection, for instance. Schools might do it, parents might do it, whether for similar, general reasons(e.g. to keep children from viewing adult material) or something more specific(parents don't want children going to a particular website). Corporations do it to try to prevent their network from being used for non-work purposes - and as they pay for the maintenance, bandwidth, electricity, and so forth, that's their prerogative. If the same restrictions let a totalitarian regime do nasty things to their people, yes, it's nasty, but it's rather hard for the software developers to prevent it while satisfying the people who want those controls closer to home.
Anything can be exploited. A chair can do a fair bit of damage to someone, but they aren't outlawed because of it. Something that makes it very easy to do harm should be controlled(a driver's license for vehicles; I personally think there should be a permit to carry a gun if there isn't already), but you can't keep people from using something good to do something bad - and some people do see the ability to e.g. censor the Internet as "something good", even though it does involve Windows.
"Guns don't kill people, people kill people" has a grain of truth to it - someone has to actually pick up that gun and fire it - but is all too often taken to excess. No, the firearm company should not be blamed - but the people distributing those guns ought to have a care who they sell guns to. Someone who mutters ominously about other guys looking at his wife while he's browsing the selection might need a small reminder about the law...
As I've said elsewhere, the official nvidia drivers caused my system to freeze and eventually reboot when I tried to load X. Nothing in the logs, they just stopped.
Most likely because it's rather hard to prove that you collected that data on your own. Logically proving a negative is extremely difficult if not impossible.
top worked for me, but I found that any other tool did not, such as gtop or the KDE usage-monitor applet included. I didn't actually get the new version installed last night(I'm going to get a network install running before I head to work) so I don't know if this has changed.
IIRC, it's noted in the handbook that if the kernel and the source tree are different versions, then such utilities as top may not properly function. It could be that these other ports haven't been re-ported to the same kernel version, and thus don't know how to get what they want from the kernel(or, depending, they may be ported to a newer kernel than the one you have installed, with the same result). I am very much a novice at programming, nowhere near the skill required to poke at the core code of my OS, so I don't know a way to verify this, much less fix it.
Hmmm, NVIDIA's driver didn't like my system, either - to the point that when X tried to start with it, my system would totally lock up. The open-source driver works much better for me.
And the logs didn't show any sort of error - everything just kinda stopped - so there wasn't much for me to submit for a bug report. Ah well, if it happens again I'll just try to be less apathetic.
Okay, Debian wasn't conflicting packages, it was KDE not running properly, outdated version of (e.g.) Gaim, getting my sound card to work...
Mandrake killed itself, Red Hat was erratic, Debian just plain got replaced.
My computer is a semi-old and quite unreliable pile o' junk. That Windows hates it is no surprise(though Windows actually does support my nVidia GeForce card's 3D acceleration). I've tried Red Hat, two versions of Mandrake, and Debian.
/dev/ums0. Much easier than muddling through a whole tangle of device nodes and hoping that one of them is what I'm looking for.
/.'d by the time I get home from work(.com is blocked by the firewall, .org is not. Maybe there's a /.er on my IT staff?).
FreeBSD is the only thing I've tried that'd keep running if I didn't poke at it. And when I did choose to poke at it, it was most tolerant of it, and - thanks in large part to the devfs system - it's FAR easier to tell what I should be poking AT. Especially for my USB card reader - attach the device, and there it is, a brand new entry,
And for all the extra time it takes, I'm very fond of the ports tree's default-ish approach of "compile from source to suit the system". My Linux experience was fraught with library conflicts in binary packages; in FreeBSD I've hit a few snags, but they were much more easily resolved - although the process was time-consuming, it was not terribly attention-consuming.
For a supposedly dead OS, FreeBSD lives quite well indeed on my system, when the Linux distros I've tried all died in short order. If only I had the space to compile OpenOffice, I'd be set.
Now I just hope the review hasn't been
I was just going to reinstall FreeBSD tonight after work. Perfect timing for a review of the new release.
Dead OS, indeed.
Even machines, though, often follow this strategy. Most dishwashers I'm aware of have a two-stage wash at least(one bin of soap being added when the door is first closed, the other some time later), and the larger washers at laundromats soak the clothes for a while before they ask for any soap to be added at all.
However, I don't know if people would be terribly impressed by a computer that, when instructed to compile, instead started trying to force-crack passwords to make sure they're secure, or some other unrelated task. Still... "make -j4" If it can't compile ONE file right away, why not work on another?