4 ms·

See also: https://mas.to/@davejmurphy/114414723608693881 https://mas.to/@davejmurphy/114414723608693881 There's a legal and moral implication. Legally, saying

by deng 1y ago

See also: https://mas.to/@davejmurphy/114414723608693881 https://mas.to/@davejmurphy/114414723608693881 There's a legal and moral implication. Legally, saying that you took someone else's code as "source of information" would probably be seen as copyright infringement. Morally, I must say I find it difficult to get worked up about it. How to write a basic OS with a threading implementation shouldn't be rocket science for anyone with a good CS background (yes, I'm aware RTEMS was literally written for rockets, but still). Yes, he took a shortcut here by looking at an existing project, and if this was a commercial project, I might see why one could get angry. However, I also can't really understand why they simply refuse to give credit to the original project and maybe also change the license to GPLv2.

msm_ 1y ago
>Legally, saying that you took someone else's code as "source of information" would probably be seen as copyright infringement Not necessarily. There are ways to work around that, for example white box reverse-engineering[1]. Also there are exemptions for some use cases, like interoperability. I'm sure a lawyer could figure something out, but it looks like they didn't consult any, and didn't follow a rigorous process. [1] https://ospo.co/blog/modern-reverse-engineering/ https://ospo.co/blog/modern-reverse-engineering/ - a random link I found by googling, but presents the idea well.
TuxSH 1y ago
IANAL, however: > maybe also change the license to GPLv2 Without a CLA, that's not doable without the approval of all the contributors. Furthermore, the point of these libraries (libogc, libctru, etc.) are to be a FOSS SDK. Non-permissive license like GPL aren't really an option in this situation. If I'm not mistaken, RTEMS has been relicensed to BSD-2; even if the licenses were incompatible at the time, they don't seem to be anymore.
- karel-3d 1y ago
  Ah then it's easy to solve, if the parties worked together.... which they won't, so, whatever
- jchw 1y ago
  > Without a CLA, that's not doable without the approval of all the contributors. Three things: - libogc is BSD 3-clause[1], which is GPL-compatible[2][3]. That means that while you can't just strip off the existing BSD license, you can distribute a combination of source code that is BSD and GPL licensed, it just has to be distributed under the terms of the GPL, which is fully compatible with the terms of a BSD license, too, since all BSD licenses require is maintaining the copyright notice/disclaimer, and nothing about that requirement conflicts with the GPL. (IANAL, so take with a grain of salt, but I'm pretty positive on this one.) - Although RTEMS is traditionally GPL, it's actually a modified version of GPL with the GNAT exception[4]. A compiler exception effectively allows users to link the object files into software of any license without invoking the copyleft provisions, so to my understanding this would basically be identical except for people who are modifying the source code of libogc (I imagine this is exceedingly rare.) - And finally, RTEMS has been re-licensing to BSD 2-clause for a long time[5], so it's very possible all of the relevant code is now available under a BSD license with effectively equivalent permissiveness. So actually I think this situation is probably much a nothingburger in that regard... [1]: https://github.com/devkitPro/libogc/blob/master/libogc_license.txt https://github.com/devkitPro/libogc/blob/master/libogc_licen... [2]: https://www.gnu.org/licenses/license-list.en.html#ModifiedBSD https://www.gnu.org/licenses/license-list.en.html#ModifiedBS... [3]: https://en.wikipedia.org/wiki/License_compatibility https://en.wikipedia.org/wiki/License_compatibility [4]: https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/LICENSE.md#rtems-license https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/LICENS... [5]: https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/LICENSE.md#bsd-2-clause https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/LICENS...

by deng 1y ago

msm_ 1y ago

>Legally, saying that you took someone else's code as "source of information" would probably be seen as copyright infringement Not necessarily. There are ways to work around that, for example white box reverse-engineering[1]. Also there are exemptions for some use cases, like interoperability. I'm sure a lawyer could figure something out, but it looks like they didn't consult any, and didn't follow a rigorous process. [1] https://ospo.co/blog/modern-reverse-engineering/ https://ospo.co/blog/modern-reverse-engineering/ - a random link I found by googling, but presents the idea well.

TuxSH 1y ago

IANAL, however: > maybe also change the license to GPLv2 Without a CLA, that's not doable without the approval of all the contributors. Furthermore, the point of these libraries (libogc, libctru, etc.) are to be a FOSS SDK. Non-permissive license like GPL aren't really an option in this situation. If I'm not mistaken, RTEMS has been relicensed to BSD-2; even if the licenses were incompatible at the time, they don't seem to be anymore.

karel-3d 1y ago
Ah then it's easy to solve, if the parties worked together.... which they won't, so, whatever
jchw 1y ago
> Without a CLA, that's not doable without the approval of all the contributors. Three things: - libogc is BSD 3-clause[1], which is GPL-compatible[2][3]. That means that while you can't just strip off the existing BSD license, you can distribute a combination of source code that is BSD and GPL licensed, it just has to be distributed under the terms of the GPL, which is fully compatible with the terms of a BSD license, too, since all BSD licenses require is maintaining the copyright notice/disclaimer, and nothing about that requirement conflicts with the GPL. (IANAL, so take with a grain of salt, but I'm pretty positive on this one.) - Although RTEMS is traditionally GPL, it's actually a modified version of GPL with the GNAT exception[4]. A compiler exception effectively allows users to link the object files into software of any license without invoking the copyleft provisions, so to my understanding this would basically be identical except for people who are modifying the source code of libogc (I imagine this is exceedingly rare.) - And finally, RTEMS has been re-licensing to BSD 2-clause for a long time[5], so it's very possible all of the relevant code is now available under a BSD license with effectively equivalent permissiveness. So actually I think this situation is probably much a nothingburger in that regard... [1]: https://github.com/devkitPro/libogc/blob/master/libogc_license.txt https://github.com/devkitPro/libogc/blob/master/libogc_licen... [2]: https://www.gnu.org/licenses/license-list.en.html#ModifiedBSD https://www.gnu.org/licenses/license-list.en.html#ModifiedBS... [3]: https://en.wikipedia.org/wiki/License_compatibility https://en.wikipedia.org/wiki/License_compatibility [4]: https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/LICENSE.md#rtems-license https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/LICENS... [5]: https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/LICENSE.md#bsd-2-clause https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/LICENS...