Linked by Thom Holwerda on Thu 8th Jan 2009 00:15 UTC
Thread beginning with comment 342800
To view parent comment, click here.
To read all comments associated with this story, please click here.
To view parent comment, click here.
To read all comments associated with this story, please click here.
RE[5]: Not a good thing when BSD thieve celebrate
by Moulinneuf on Fri 9th Jan 2009 13:31
in reply to "RE[4]: Not a good thing when BSD thieve celebrate"
RE[6]: Not a good thing when BSD thieve celebrate
by sean on Fri 9th Jan 2009 18:36
in reply to "RE[5]: Not a good thing when BSD thieve celebrate"
GPLv2 is compatible with GPLv3.
Only if the GPLv2 license carries the "or later" clause. For example, the Linux kernel does not, therefore, it is not compatible.
The hurd is not a failure , it's not BSD's , people can still develop it.
Huh? Both the BSD's and HURD can still be developed. HURD is not as successful as Linux nor the main BSD OS's.
RE[5]: Not a good thing when BSD thieve celebrate
by segedunum on Sat 10th Jan 2009 16:24
in reply to "RE[4]: Not a good thing when BSD thieve celebrate"
There is another reason: compatibility, the more permissive license you use, the less license incompatibility mess there is
License compatibility is a fairly weak argument for using a license, and it isn't going to make your software better or encourage more contributions.
As for the proprietarisation of MIT/BSD code, this doesn't always happen: the OpenBSD and FreeBSD OS seems to have a working ecosystem, sure they're less popular than Linux, but is the license the reason?
Given that BSD in one form or another has been around for longer than Linux, you'd have to say yes. The GPL makes everyone contributing to the Linux kernel know that they're on a level playing field. If you contribute something then even your competitors need to contribute back if they want to have any influence. This takes any political guesswork out of contributing. Well, most of it anyway.
With a BSD (or more 'liberal') licensed kernel is a hardware manufacturer going to work on open source drivers, or contribute in a big way to major kernel subsystems when a rival can just plug in a proprietary extension and sell that? I very much doubt it. Without the GPL it is very much a "Who is going to blink first?" scenario.
It's not possible to isolate the reason: The Hurd was a failure even though it is GPL licensed..
Hurd has become unnecessary because people were already using something else at its inception and it isn't inherently better in any way to anything else. No license can help you with that.






Member since:
2005-07-06
[[The number one reason why more permissive licenses like MIT, BSD or even the LGPL tend to be used is because of linking to proprietary software. There really isn't any other reason you can come up with.]]
There is another reason: compatibility, the more permissive license you use, the less license incompatibility mess there is (GPLv2 isn't even compatible with GPLv3!).
As for the proprietarisation of MIT/BSD code, this doesn't always happen: the OpenBSD and FreeBSD OS seems to have a working ecosystem, sure they're less popular than Linux, but is the license the reason?
It's not possible to isolate the reason: The Hurd was a failure even though it is GPL licensed..