[Qgis-psc] QGIS licence (was: QGIS on iOS was QGIS for Mobile (Android))

Thu Oct 17 03:04:02 PDT 2019

Hi all

Let me share my thoughts...

- re Ars Technica article - I agree with Andreas we are in different
position, having a great community project with no single company
owning it. Anyway the article is a good reminder of possible threats
that FOSS projects might face (there's also an interesting FOSDEM
keynote [1] about that). Even though right now it concerns mostly
server-side software, I can see also a possibility that in future
people would be buying cheap computers with closed operating systems
that would only run cloud-hosted desktop apps - and people would have
to pay for running the apps even though the apps were open source
licensed.

- possible new license - just to make sure we are on the same page -
are we discussing just the step to make the license less restrictive
or also an opinion to make it more restrictive? I believe we were only
considering the option for less restrictive license...?

- I would prefer if we can make a summary of what would we expect from
the license change - what are our motivations, what we don't care
about, and what we definitely don't want - this could give an
indication how much consensus there is in the community.

- my personal take on the possible license change is that I would
welcome fewer license restrictions on QGIS *libraries*. For example,
sometimes I would love to be able to use qgis_core library for some
work - but GPL requires that the derived work is also licensed in a
similar way, which is not always possible (due to client's
requirements). So I would end up using other, less advanced tools for
the task. Having a lot of experience with QGIS API, this is pity
because I could spend less time coding boring bits (which are
available in QGIS libraries) and focus more on making QGIS libraries
even more useful.

- I like Nyall's idea of getting some impartial advice on the matter
of a possible license change - the question is who would consider
themselves as unbiased enough :-)

Cheers
Martin

[1] https://www.youtube.com/watch?v=goQ-yjN7WAc

On Thu, Oct 17, 2019 at 11:09 AM Andreas Neumann <andreas at qgis.org> wrote:
>
> Hi Nyall,
>
> I had a look at the article at ars-technica. It mainly discusses the issue where other companies use and resell OS software in their cloud environment and don't give anything back to the OS project, esp. the cloud providers like AWS, Azure, Google, etc. This seems to be esp. difficult for server-side SaaS applications (example MongoDB). However, QGIS is still mostly a Desktop application, though a lot of effort and investments also go into the server side and mobile versions. If QGIS server got very, very popular one day in the future, it might be a problem. I wish that QGIS server would be so important or well-known that AWS, Azure or Google would offer it as a managed service ...
>
> I think you are right that these issues should be discussed. But I am not convinced the MongoDB Horowitz approach to restrict usage in a cloud environment is the right way to go. Switching to a more permissive license than GPL (e.g. to make it compatible with the Mac app store) makes the problem even worse in the cloud environment.
>
> Nyall: Do you already see examples where the investments/business of the QGIS core contributor companies are under pressure by QGIS being GPL licensed? Aren't those core QGIS companies mainly doing business in development and support? I don't see how a switch to a different license would better protect their work or business around QGIS. But I haven't dived in to this issue more. If you can give more examples / thoughts were a different license could improve things, it would help.
>
> If QGIS would switch to a more restrictive license in terms of what you can do with it (like MongoDB Horowitz suggests) than maybe come QGIS companies would attract more money, but QGIS as a project would probably risk loosing a lot of contributors.
>
> Thanks,
> Andreas
>
> On Thu, 17 Oct 2019 at 08:57, Nyall Dawson <nyall.dawson at gmail.com> wrote:
>>
>> On Thu, 17 Oct 2019 at 16:42, Andreas Neumann <a.neumann at carto.net> wrote:
>> >
>> > Hi Nyall,
>> >
>> > I doubt that an "unbiased" company/organization/person even exists on this planet, when it comes to licenses, just as people have a precedence for political views and parties. Disclaimer: I am personally certainly not unbiased, regarding this matter.
>> >
>> > But before even starting on asking someone to dive into the topic, I would like to know the reasons why we need to discuss a license change?
>> >
>> > Is iOS the only reason? If yes, others have pointed out that alternative app stores exist. If we offer QGIS on iOS, we could publish it in an alternative app store, write some small page explaining how it works and the problem should be solved? It is certainly acceptable for someone to do these 2-3 extra clicks in order to get a free QGIS mobile for their Apple mobile? I am sure our users would understand this situation, if we explain in 2-3 sentences why we have to do this: license incompatibilities that aren't easy (or maybe not even possible?) to solve.
>>
>> While it's been the trigger of the current round of discussion, It's
>> certainly not the only factor at play. For instance, I know the GPL
>> requirement on QGIS plugins has been a sticking point for some in the
>> past.
>>
>> I think there's also a need to discuss licensing from a
>> risk-management perspective.There's a reasonable post published
>> yesterday on this topic over at ars technica:
>> https://arstechnica.com/information-technology/2019/10/is-the-software-world-taking-too-much-from-the-open-source-community/,
>> which covers some of the same territory as Paul Ramsey's recent
>> keynotes have done (but arguably Paul's versions are better researched
>> and more eloquent). The face of open-source development IS changing,
>> and the software world is HUGELY different today vs what it was when
>> the QGIS project began. I honestly feel that it's really just doing
>> due-diligence for us as a project to at least investigate these
>> different factors and have a trustworthy, well researched body of
>> knowledge surrounding it. A large percentage of this community have
>> livelihoods/incomes of which QGIS is a significant component, and I'd
>> really like to see us as a project recognise and take steps to protect
>> this.
>>
>> At the moment, we're all just throwing around personal opinions
>> (myself included) based on incomplete understanding of the whole
>> situation...
>>
>> Nyall
>>
>>
>>
>>
>> >
>> > Andreas
>> >
>> > On 2019-10-17 08:11, Nyall Dawson wrote:
>> >
>> > On Thu, 17 Oct 2019 at 15:44, Paolo Cavallini <cavallini at faunalia.it> wrote:
>> >
>> > no need to be emotional on this, fully agreed.
>> > So if I if I understand it correctly you are suggesting to move to a
>> > BSD-like licence or an LGPL for part of it?
>> >
>> >
>> > Actually - I'm proposing a discussion several levels before even
>> > discussing a particular license :)
>> >
>> > I'd like someone to research and write up a commissioned report on whether:
>> >
>> > - it would be possible to relicense QGIS and what would be involved in
>> > doing so (including which other projects have done this and how they
>> > went about it)
>> > - what the potential benefits and downsides of doing so would be.
>> > - possible licenses we could investigate, and which of these would
>> > make things like the iOS situation easier
>> > - whether dependencies we already have would block any possibility of
>> > relicensing
>> > - how this would impact on the plugin scene (I could see there being
>> > both huge benefits and disadvantages of relicensing for plugins)
>> > - whether a potential CLA assigning code ownership to qgis.org would
>> > be possible, and the advantages and disadvantages of this
>> >
>> > I think this should be done by a trusted company or individual, and
>> > ideally one well acquainted with the QGIS community yet with a history
>> > of **impartiality** to topics which impact on this (like the ios
>> > situation, or a history of antagonism toward non-GPL software). Only
>> > after a report has been written by this individual/organisation, and
>> > then tabled and accepted by the PSC, should it THEN be raised for
>> > public discussion with the community, where **everyone** discussing
>> > the issue can be fully informed of all sides of the debate **before**
>> > the group discussion even begins!
>> >
>> > Nyall
>> >
>> > Besides personal preferences
>> > and priorities, I believe this would be technically very difficult. We
>> > tried many years ago the same for GRASS, and it proved impossible,
>> > especially because of the code from developers disappeared form the radar.
>> > I should add that I'm hearing since years rumors of the type "it's free
>> > for now, be sure that when it will be very good you'll have to pay for
>> > it". I think that discussing about this will give credit to these
>> > rumors, and spread FUD around the project, so we have to be extra careful.
>> > I suggest discussing this in the next PSC.
>> > Cheers.
>> > --
>> > Paolo Cavallini - www.faunalia.eu
>> > QGIS.ORG Chair:
>> > http://planet.qgis.org/planet/user/28/tag/qgis%20board/
>> > _______________________________________________
>> > Qgis-psc mailing list
>> > Qgis-psc at lists.osgeo.org
>> > https://lists.osgeo.org/mailman/listinfo/qgis-psc
>> >
>> > _______________________________________________
>> > Qgis-psc mailing list
>> > Qgis-psc at lists.osgeo.org
>> > https://lists.osgeo.org/mailman/listinfo/qgis-psc
>> >
>> >
>> _______________________________________________
>> Qgis-psc mailing list
>> Qgis-psc at lists.osgeo.org
>> https://lists.osgeo.org/mailman/listinfo/qgis-psc
>
>
>
> --
>
> --
> Andreas Neumann
> QGIS.ORG board member (treasurer)
> _______________________________________________
> Qgis-psc mailing list
> Qgis-psc at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/qgis-psc