Re: The sad decline of copyleft software licenses? :(

[Pen-Yuan Hsing]

On 9/25/20 10:12 PM, Jean Louis wrote:
> * Pen-Yuan Hsing <penyuanhsing@gmail.com> [2020-09-25 04:08]:
> > ... a big
> > majority of new software published on platforms such as GitLab or GitHub
> > uses permissive licenses with MIT being one of the most popular. I
> > distinctly remember a graph published, probably by GitHub, or the proportion
> > of different licenses used and the graph shows this trend. (sorry I can't
> > find the original graph right now)
>
> Maybe it is this one:
> https://github.blog/2015-03-09-open-source-license-usage-on-github-com/

That could very well be it, thank you for finding it! While there is a 
substantial number of copyleft-licensed software shown in that graph, 
they are in a clear minority.

> Their wordings for the MIT license is:
>
> "The MIT License is short and to the point. It lets people do almost
> anything they want with your project, like making and distributing
> closed source versions."
>
> Their wording is not quite nice, as making and distributing "closed
> source" or binary versions is allowed with GPL, what is not allowed is
> disallowing the user to receive the source code.

I agree the wording can be made more precise. And the "short and to the 
point... lets people do almost anything they want with your project" 
implies that this is the easiest set-and-forget license to choose. For 
those who only think about open source as a development method and/or 
haven't thought deeply about software freedom, it is easy to choose the 
MIT license by default.

> Instead of thinking that MIT is competition for GPL, simply count how
> many new software is published under GPL, yesterday I have noticed
> that this game is under GPLv3:
> https://frictionalgames.com/2020-09-amnesia-is-now-open-source/

I haven't played that game before, but incidentally I also saw that 
announcement just before your email.

> So isn't that great news?

Of course it is great news.

> Instead of looking at MIT licensed software as competitor's product,
> look at it as a friendly and free software product, as the MIT license
> is free software license, appreciate it for free software production
> and push the idea and implementation of more GPL software.

I am sorry if my original post mislead you, but I never claimed that 
MIT-licensed software is "competition" or not "friendly". They are of 
course free as in freedom software and that is itself a good thing.

The point I am trying to make is that, yes, of course "permissive" 
licenses and copyleft licenses both apply to free software. However, 
permissively-licensed free software has a higher risk of being 
incorporated into proprietary software, and in my opinion that is a 
highly undesirable outcome. Because of that, I believe copyleft licenses 
are better are protecting user freedoms. Again, this does not malign 
permissive free software licenses.

As discussed elsewhere such as in Aaron Wolf's response, it would be 
good for us to communicate the social benefits of copyleft and encourage 
its adoption. Doing so *does not* have to make permissive licenses or 
its advocates enemies. I see it more as introducing an even better 
solution (i.e. copyleft license) to friends of free software.