[riot-users] [riot-devel] Switch to BSD?

Emmanuel Baccelli Emmanuel.Baccelli at inria.fr
Thu Dec 4 17:42:57 CET 2014


Hi again everyone,

If we're going for a license that is less restrictive than LGPL, it does
indeed make sense to consider alternatives such as Apache and MIT -- and
not only BSD.

However, I think we need to distinguish between two aspects:
- the ideas
- the effects in practice

The reason we consider moving away from LGPL is because, although the
*idea* of GPL is great, its *effect* is a problem for us in practice.
Experience so far shows LGPL is stopping RIOT adoption for too many
companies we interact with.

So where I'm getting at is: we need to make sure that we don't fall into
the same trap with a new license. For example: if the idea of Apache seems
better to us, but it turns out that in practice most companies we interact
with do not want to read the term "patent" in the license, we're back to
square one, which may or may not be what we want [1]. Let's assume it's not
what we want.

For the moment, surprisingly, I have not heard any company rant against BSD
-- quite the contrary in fact. So while BSD may be considered old and
incomplete, it is still a strong contender in my opinion.

Personally, I agree that the idea of the Apache license seems better,
indeed. But let's check what IoT company (lawyers) think about the Apache
license. I think we have to assess whether or not Apache is, in our case,
yet another "good idea" which turns out to be a show-stopper in practice.

Caveat: IANAL  -- a funnily appropriate acronym I just saw in a related
post [2]. If anyone has professional or personal ties with an IoT company
lawyer, please do not hesitate to chip in at this point.

Cheers,

Emmanuel


[1] Note that we could also choose to stay in square one, and dedicate
time/energy to fight for the idea. But in my opinion, we have to choose our
battles: our plates are already full with other things that none other than
us fight for (i.e. RIOT adoption & quality IoT code ;).

[2]
http://programmers.stackexchange.com/questions/40561/is-bsd-license-compatible-with-apache

On Thu, Dec 4, 2014 at 5:05 PM, Emmanuel Baccelli <
Emmanuel.Baccelli at inria.fr> wrote:

> Hi again everyone,
>
> If we're going for a license that is less restrictive than LGPL, it does
> indeed make sense to consider alternatives such as Apache and MIT -- and
> not only BSD.
>
> However, I think we need to distinguish between two aspects:
> - the ideas
> - the effects in practice
>
> The reason we consider moving away from LGPL is because, although the
> *idea* of GPL is great, its *effect* is a problem for us in practice.
> Experience so far shows LGPL is stopping RIOT adoption for too many
> companies we interact with.
>
> So where I'm getting at is: we need to make sure that we don't fall into
> the same trap with a new license. For example: if the idea of Apache seems
> better to us, but it turns out that in practice most companies we interact
> with do not want to read the term "patent" in the license, we're back to
> square one, which may or may not be what we want [1]. Let's assume it's not
> what we want.
>
> For the moment, surprisingly, I have not heard any company rant against
> BSD -- quite the contrary in fact. So while BSD may be considered old and
> incomplete, it is still a strong contender in my opinion.
>
> Personally, I agree that the idea of the Apache license seems better,
> indeed. But let's check what IoT company (lawyers) think about the Apache
> license. I think we have to assess whether or not Apache is, in our case,
> yet another "good idea" which turns out to be a show-stopper in practice.
>
> Caveat: IANAL  -- a funnily appropriate acronym I just saw in a related
> post [2]. If anyone has professional or personal ties with an IoT company
> lawyer, please do not hesitate to chip in at this point.
>
> Cheers,
>
> Emmanuel
>
>
> [1] Note that we could also choose to stay in square one, and dedicate
> time/energy to fight for the idea. But in my opinion, we have to choose our
> battles: our plates are already full with other things that none other than
> us fight for (i.e. RIOT adoption & quality IoT code ;).
>
> [2]
> http://programmers.stackexchange.com/questions/40561/is-bsd-license-compatible-with-apache
>
> On Thu, Dec 4, 2014 at 9:29 AM, Joakim Gebart <joakim.gebart at eistec.se>
> wrote:
>
>> I am also very much in favor of using a license which requires
>> openness but like Adam said, in the embedded world it is quite common
>> that changes will be necessary in order to support some hardware
>> configuration. Additionally, the interpretation that we would need
>> dynamic linking in order to comply with the license without opening up
>> all application code makes this a quite important question.
>>
>> Companies are not always willing or even able (because of patents,
>> NDAs or other contracts) to release the source of proprietary
>> applications. The use of LGPL in RIOT has been the source of some
>> discussion between me and my colleagues and I hope to see some other
>> license in the future where it is possible to still distribute
>> proprietary applications that run on RIOT.
>>
>> Eistec (see www.eistec.se) generally has the policy that anything
>> related to the platform and OS (cpu drivers, device drivers, etc) will
>> be sent upstream to related OSS projects, mainly RIOT and Contiki for
>> now (but we have also provided some patches for other tools we use,
>> such as OpenOCD), but we usually want to keep application code
>> (algorithms, higher level service implementations etc.) proprietary.
>> Since we work as a consulting firm it is also common that we do not
>> own the code to the applications themselves but have to negotiate with
>> the client on what parts to release, clients are usually fine with
>> sharing bug fixes and low level driver and OS code with upstream.
>> So far we have only used Contiki commercially but I personally would
>> like to see that change in the future, but for now I think the risk of
>> ending up in a situation where someone can demand any proprietary
>> application code from us makes this a bit too dangerous.
>>
>> This is my personal view on the license situation, but I know that
>> many of the people I work with share this concern.
>>
>> Best regards
>> Joakim Gebart
>> Managing Director
>> Eistec AB
>>
>> Aurorum 1C
>> 977 75 Luleå
>> Tel: +46(0)730-65 13 83
>> joakim.gebart at eistec.se
>> www.eistec.se
>>
>>
>> On Thu, Dec 4, 2014 at 5:06 AM, Adam Hunt <voxadam at gmail.com> wrote:
>> > While I've been a fervent supporter of the GPL for many years I'm on
>> board
>> > with a change to a simple BSD or MIT style license. Initially I was
>> > skeptical about the need to move away from the LGPL but in the world of
>> > embedded systems it's very common to make changes to the core codebase
>> in
>> > order to work on various platforms. Under the LGPL such changes would
>> have
>> > to be tracked, checked for IP conflicts, and made available. This
>> > requirement may very well end up being so onerous that it may vary well
>> push
>> > companies to adopt a more suitably licensed alternative over RIOT.
>> >
>> > On Wed, Dec 3, 2014, 2:14 PM Thomas Watteyne <
>> watteyne at eecs.berkeley.edu>
>> > wrote:
>> >>
>> >> Emmanuel,
>> >> I support the change to BSD. One of the reasons is that OpenWSN is
>> also on
>> >> BSD, so integration of the different code bases might be easier when
>> both
>> >> have the same license.
>> >> Thomas
>> >>
>> >> On Wednesday, December 3, 2014, Emmanuel Baccelli
>> >> <Emmanuel.Baccelli at inria.fr> wrote:
>> >>>
>> >>> Dear RIOTers,
>> >>>
>> >>> we have been receiving an increasing amount of negative feedback from
>> >>> various companies concerning the practical usability of our LGPL
>> license in
>> >>> their context, being a show-stopper.
>> >>>
>> >>> For this reason, INRIA, Freie Universitaet (FU) Berlin and Hamburg
>> >>> University of Applied Science (HAW) are currently considering
>> changing the
>> >>> license of their contributions to RIOT to a less restrictive license
>> (i.e.
>> >>> BSD, potentially as soon as next release).
>> >>>
>> >>> Such a switch to BSD is betting that the effect of a potentially
>> smaller
>> >>> percentage of user/devel contributing back to the master branch will
>> be
>> >>> dwarfed by the effect of a user/devel community growing much bigger
>> and
>> >>> quicker. This seems doable considering the current momentum around
>> RIOT.
>> >>>
>> >>> In a second phase, if such a license switch takes place for
>> INRIA/FU/HAW
>> >>> contributions, we would then contact other contributors individually,
>> to
>> >>> check their status concerning a similar switch for their own
>> contributions.
>> >>>
>> >>> But in the first place, we would like to debate this topic. In
>> >>> particular: is anyone violently opposing the idea of migrating to a
>> less
>> >>> restrictive license, such as BSD? If so, why? On the other hand, if
>> you
>> >>> explicitly support the license change, feel free to indicate this as
>> well.
>> >>> Please send your opinion to the list before Dec. 10th.
>> >>>
>> >>> Cheers,
>> >>>
>> >>> Emmanuel
>> >>
>> >> _______________________________________________
>> >> devel mailing list
>> >> devel at riot-os.org
>> >> http://lists.riot-os.org/mailman/listinfo/devel
>> >
>> >
>> > _______________________________________________
>> > devel mailing list
>> > devel at riot-os.org
>> > http://lists.riot-os.org/mailman/listinfo/devel
>> >
>> _______________________________________________
>> devel mailing list
>> devel at riot-os.org
>> http://lists.riot-os.org/mailman/listinfo/devel
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.riot-os.org/pipermail/users/attachments/20141204/a4a99f4b/attachment.html>


More information about the users mailing list