LibertyBSD

6 replies [Last post]
loldier
Offline
Joined: 02/17/2016

What about LibertBSD? Still active?

https://libertybsd.net/

The adjacent forum seems to have disappeared. JadedCtrl no longer posts anything here, either.

SuperTramp83

I am a translator!

Offline
Joined: 10/31/2014

Jaded does post but rarely. You can find her on IRC, freenode, #lgn.

mao

gd_scania
Offline
Joined: 09/13/2017

I think LibertyBSD will be needed to switch to the FreeBSD base system w/o ZFS support, just using the GNU+HurdFS and UFS for installations.
I also wonder that, why the license behind ZFS, CDDL, is GPL-violated, and what's made to a non-copyleft free software license GPL-compatible or GPL-violated?

Magic Banana

I am a member!

I am a translator!

Offline
Joined: 07/24/2010

Copylefted licenses usually state that derivative works must be under the same license than the original work. That is why two pieces of code under two different copylefted licenses cannot be combined: the resulting combination would have to be under both licenses at the same time, what is impossible.

strypey
Offline
Joined: 05/14/2015

MagicBanana:
"two pieces of code under two different copylefted licenses cannot be combined: the resulting combination would have to be under both licenses at the same time, what is impossible."

IANAL but doesn't that depend on whether the licenses in question contradict each other? If not, wouldn't it be possible to dual-license, as long as you can satisfy the conditions required by both licenses?

Some illuminating discussion on the alleged incompatibility between GPL and CDDL can be found here:
https://opensource.stackexchange.com/questions/2094/are-cddl-and-gpl-really-incompatible

Magic Banana

I am a member!

I am a translator!

Offline
Joined: 07/24/2010

I do not find the most voted answer (probably the one you refer to) illuminating: it pretends that section 0 of the GPL makes ambiguous claims and that it enforces things not explicitly listed in the US copyright law. Those are the far-from-clear "arguments" (what claims? what things?) that lead to: "As a result, both jurisdictions make the claims in GPL section 0 (where the GPL tries to defined what a derivative work is) void claims".

The inventor of the copyleft and main author of the GPL, RMS, disagrees (in an article that is more recent than the comment you refer to):

With static linking, it is self-evident what code is part of the source code for that entire program. It includes, at least, the source code for all modules linked together in the binary.
Developers often find this point not quite so self-evident with dynamic linking, but the situation is equally clear: if you distribute modules meant to be linked together by the user, you have made them into a combined work, and you must release the entire combined work under the GNU GPL.
In other words: dynamic vs. static linking never makes any difference on the outcome of the analysis. The FSF has been advised by several US lawyers on this matter over the years, and the answer is always this one. Unsurprisingly, some lawyers have been willing to defend a different interpretation when it suits their clients, but their arguments are weak and disputed by the majority of experts.
https://www.fsf.org/licensing/zfs-and-linux

Not only the FSF claims that the GPL is incompatible with the CDDL but the authors of the CDDL too:

Like the MPL, the CDDL is not expected to be compatible with the GPL, since it contains requirements that are not in the GPL (for example, the "patent peace" provision in section 6). Thus, it is likely that files released under the CDDL will not be able to be combined with files released under the GPL to create a larger program.
https://lwn.net/Articles/114840/

In fact, the main author of the CDDL, Danese Cooper, said the CDDL was based on the MPLv1.1 partly because the MPLv1.1 is incompatible with the GPL:

Mozilla was selected partially because it is GPL incompatible. That was part of the design when they released OpenSolaris.
http://meetings-archive.debian.net/pub/debian-meetings/2006/debconf6/theora-small/2006-05-14/tower/OpenSolaris_Java_and_Debian-Simon_Phipps__Alvaro_Lopez_Ortega.ogg at 27:27

jxself
Offline
Joined: 09/13/2010

"IANAL but doesn't that depend on whether the licenses in question contradict each other?"

...and copyleft licenses usually do for the reasons Banana explained: You have two licenses both in a virtual tug-of-war saying "you must put the combined whole under ME... no ME... no ME... no ME..." *KA-BOOM* (I say "usually" solely because some will have compatibility terms inside of them with specific licenses like the compatibility clause in both GPLv3 and AGPLv3 that makes them compatible with each other and solves the tug-of-war. But most don't have this.)

"If not, wouldn't it be possible to dual-license, as long as you can satisfy the conditions required by both licenses?"

A second license would need the consent of all copyright holders. In large projects like the kernel named Linux individual contributors retain their own copyright and there is no central authority with power to either relicense the project or add a second one. So the first step is to get the permission of every copyright holder. With the kernel named Linux having many thousands of copyright holders spanning nearly 27 years finding all of them (or their heirs if they're dead) and then getting all of them to agree to adding a second (not GPL) license for compatibility with ZFS can prove to be impractical.

And if you want to try it from the other side of changing the license on ZFS you have a similar problem there with tracking down all of the contributors and getting all of them to agree. And one of those will be Oracle. The free software community does not typically look to Oracle for pro-freedom activities. Good luck getting them to agree.