Re: Three problems with Chinput
Hi
It will violate GPL modules in kernel. When you load NVidia module, the kernel
will warn you. ;-)
--
Yu Guanghui <ygh at dlut.edu.cn>
Network Center
Dalian University of Technology, China
引用 James Su <suzhe@gnuchina.org>:
> Then, how do you think about the relationship between the linux kernel
> and close sourced modules, like NVidia kernel driver? Don't they obay
> those two rules?
>
> Regards
> James Su
>
>
> Yong Li wrote:
>
> >On Thu, 1 May 2003, Yu Guanghui wrote:
> >
> >
> >
> >>Hi
> >> I didn't see any violation. SCIM is LGPL, the upsteam author (Su Zhe)
> provides
> >>the source. The upstream author also writes the pinyin module, and decides
> not
> >>to release the source. Pinyin module should not be considered as part of
> SCIM.
> >>It's a standlone program use SCIM API that releases under LGPL.
> >> So the SCIM will be put in Debian main (it's LGPL and don't depend on
> pinyin
> >>module, there are other input modules), and pinyin will be put to
> non-free.
> >>
> >>
> >
> >Here is why I think it might be a violation.
> >
> >As I understand it scim is under LGPL and pinyin is a loadable module. For
> >this combination I'm only aware of two possibilities for the module not
> >have to be released under LGPL or GPL-compatible license:
> >
> >1. If the aforementioned module is a standalone executable and only
> >dynamically linked to the libraries of scim. when started the module must
> >run in its own space.
> >
> >2. If the module is not a standalone executable and is dynamically loaded
> by
> >the main program (scim in this case). Then the main program must use fork
> >and exec to invoke the module to make it a separate program.
> >
> >The pinyin module is obviously not a standalone executable! It is
> >physically distributed as a so file. So possibility 1 is definitely out.
> >All the evidences suggest that scim doesn't fork and exec the module.
> >Instead it is loaded in its entirety into the main program's space.
> >Function calls are made and data structures are shared between the two
> >entities. So, imho, they form a single program. As such the module must be
> >treated as an extension to the main program and be released under the LGPL
> >or a GPL-compatible license. You might want to refer to the topic on how
> >to treat the plug-ins for a GPLed program in GNU's GPL FAQ.
> >
> >The uniqueness about this scim situation is that it is the author himself
> >who "violates" his own license. For all the years I have been following
> >the open source development, this is the first time I have ever seen this
> >happen. Strictly speaking it is not a violation for the author himself.
> >Because from legal standpoint of view, a license such as GPL/LGPL is a
> >contract from the author for the users of his program. The author himself
> >is not bound by the license. However I think for everyone else, such as
> >Anthony, who want to redistribute the pinyin module, it is a violation of
> >the LGPL license.
> >
> >Of course this is only my personal opinion based on my own understanding
> >of GPL/LGPL license. This is such a rare and unique case. That's why I
> >think it's important, if you want to include it in debian distribution, to
> >bring this up to debian-legal and hear what those experts have to say.
> >
> >Granted even if someone redistributes the pinyin module and violates the
> >LGPL license, it's highly unlikely the author will ever pursue him/her and
>
> >enforce the license. After all it's the author's intension of doing so.
> >So the legal risk is minimum. However this does not make it right. If this
>
> >is indeed a LGPL violation, I think it's a shame for the debian project to
>
> >put it into its distribution.
> >
> >Regards,
> >rigel
> >
> >
> >
> >
> >
> >
> >
> >
>
>
>
>
-------------------------------------------------
歡迎使用大連理工大學web郵件系統: http://mail.dlut.edu.cn
--
To UNSUBSCRIBE, email to debian-chinese-gb-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
--
| This message was re-posted from debian-chinese-gb@lists.debian.org
| and converted from gb2312 to big5 by an automatic gateway.
Reply to: