On Tue, Jun 16, 2020 at 09:28:05AM +0100, Michael Tremer wrote:


> I consider myself a great advocate for free software. Almost everything I do,
> and certain all I can, is free software - available for anyone to use.

Let me nitpick on that Free software also requires the ability to modify and
distribute the modified work…

> We have spent a lot of time on this and we do not want another Maxmind. I am
> not trying to make money with this project, but nobody else should be making
> that money either.

Being "gratis" is stricly not a requirement for FLOSS*, but being "libre" is,
and the 4 software freedoms encourage to allow usage for "any purpose", 
commercial use. So, IMHO, this two paragraphs are somehow conflating gratis
with libre. (Additionally CC-BY-SA does not have a commercial-usage restriction,
as some said already in this thread)

* for example, there are some projects (in the Android App world) that sells
the app in the offical store but have the sourcecode available to compile
yourself on a public repository)

> Since this is only a license - and people seem to rather ignore than follow
> these - there is no guarantee for us that someone does things that we do not
> want them to do. But in the end I have to protect my project and the other
> people working on this so that we can continue doing this.
> I do not want this to be non-free, but I hope my point makes at least some
> sense.

if you want it to be (DSFG)-free, please choose one of the approved licenses.
But I fear that your expectations are different: A (DFSG-)free license must
not limit commercial use in any way, for example…

> >> * it would be nice to encourage users to give back to the project and help
> >> them to help us to improve the data wherever possible
> > 
> > Such encouragements should be part of, e.g., a README file, but not part of
> > a license. *Forcing* users to contribute back would likewise make a license
> > non-free for Debian usage (since that would fail the Desert Island test).
> Sorry for my noob question, but doesn’t the GPL “force” people to give back?

No, it does not. We had some discussion about a different license that crossed
this topic lately: https://lists.debian.org/debian-legal/2020/04/msg00016.html
TL;DR: You only need to offer your modifications to _your_ recipients of the
work, not to whom you received it from.

> > Fortunately, you said "encourage", so that would be optional and hence
> > good. I'm just pointing out that even ideas with good intentions (naturally
> > improving a database is a plus for any user) can lead to software or data
> > becoming non-free.
> A license is just letters on some paper. I had my own software copied too
> often by too many people with bad intentions and I could not do anything
> about it without throwing more money and time down the drain.

There is this famous "use if for good not evil"-Json-license … disaster? …
A true free software must even allow usage for evil purposes, not even touching
the question who defines "evil"?

> So, I guess we can conclude that the CC BY-SA 4.0 option is definitely
> something that we would drop. Simply for that reason that it is too
> complicated.

Chooose any license you find suitable. Best from 
Some people will disagree on some license being listed here*, but this is kind 
official position of the project. (there are sometime more factors than a
license to consider something free software) The only strong advice I would
like to give is "Don't invent your own licence."

*For instance, I'm not in the CC-4.0-is-non-free camp, but I would love to
learn about the objections…

> I always assumed that any of the GPL licenses won’t be applicable to data
> (and only code). Can maybe brings some light into the dark for me?

Many people believe that can be applied to data as well, incl. the FSF [1].
/me has e.g released CAD models [2] using the GPL, but I explicitly
clarified that I consider this covered, no idea if that would be actual
needed, though. IANAL.

[1] https://www.gnu.org/licenses/gpl-faq.en.html#GPLOtherThanSoftware
[2] e.g 


Reply via email to