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

[Jean Louis]

* Pen-Yuan Hsing <penyuanhsing@gmail.com> [2020-09-28 05:48]:
> > I cannot see there possible violation of the GPL.
> > 
> > You can use R and make R programs which are proprietary. You can have
> > bash shell and have shell scripts which are proprietary. You can have
> > proprietary program producing automatically shell scripts executed by
> > shell, whereby such shell scripts would be proprietary. I do not vouch
> > for proprietary software, as free software has a lot to do in the
> > world. But that is how it is, the GPL license have been designed in
> > such a way to allow such usages of software.
> 
> I agree with the points you made. However, I must apologise if I mislead
> you, because I *did not* claim that the proprietary software Brodgar
> violates the GNU GPL of R. Like I said in the example, the Brodgar homepage
> specifically has a highly visible section telling you why it does not
> violate the GPL. The point I am trying to make with this example is not of a
> GPL violation. The bigger context of my argument is that copyleft licenses
> are better than permissive licenses, because permissively-licensed free
> software has a higher risk of being used in proprietary software. The
> Brodgar example is to illustrate that even with a copyleft license such as
> the GPL, there is still a way for GPL-licensed free software to be used with
> proprietary programs without violating the GPL. If that can be done, then it
> is *even easier* for permissively-licensed free software to be exploited by
> proprietary software. The whole example was meant as an argument that
> copyleft licenses are at least better than permissive ones, even if copyleft
> is not perfect and might have loopholes.

Sure that GPL program may be invoked automatically by proprietary
program.

But that capability is not abuse, and never was. It is feature of the
GPL, you may use the program how you wish, you see? As a feature, that
should be benefit, not disadvantage.

GPL program may be used for crime, smuggling, nuclear missiles, drug
sales, and other destructive actions. But that is still a feature of
free software, and not disadvantage. You may use it for whatever
purpose you wish, that includes invoking it by proprietary
software.

> Denver Gingerich's response suggested that it is entirely possible Whatsapp
> negotiated a custom license from Open Whisper Systems to use the Signal
> implementation to bypass the GPL altogether.

Well, in their documents they say they use GPL, maybe among millions
of users, nobody asked for the sources.

> > I do not share opinion that they are even more dangerous from
> > perspective of user freedom.
> 
> Are you referring people or licenses not being more dangerous?

Licenses.

Because permissive license already gave source code and permission for
you to use the free software. From perspective of user freedom, users
got freedom that way. 

> I don't understand what you're trying to communicate here. I think it's
> pretty clear that the creation and spread of proprietary software would be
> easier if free software uses permissive licenses instead of copyleft
> licenses.

I am saying it does not matter, you are getting free software
too. OpenBSD is mostly licensed that way, but I enjoy the free
software as OpenBSD, not proprietary. It does not matter.

Finally, GPL software can also be made to proprietary, which does
happen, so it is violation, you cannot practically do much, but it can
be made into proprietary. License alone is preventing only ethical
software authors not to abuse it, those not ethical will abuse it
anyway.

Making any proprietary software is abuse. Yet license alone will not
stop companies abusing it, be it GPL or MIT-similar license. I guess
there is no such thing any more as MIT license, it is better to say
MIT-similar or MIT-like.

It is like guns, you need a license for gun, but bad boy need not
license, is free to use a gun in any country.

> I have never claimed that those who apply permissive free software licenses
> to their code are competitors nor was I implying that.
> 
> However, I am arguing that once a person creates a piece of software, and if
> that person cares about perpetuating user freedom, choosing a copyleft
> license would be a better way to achieve that goal. That's why I find the
> decline in usage of copyleft licenses a sad thing.

OpenBSD developers and FreeBSD developers, and many others, don't
share that opinion, it is thus better working together in creation of
free software.

I do not see decline of copyleft licenses, and I have given you
example that you are looking into wrong statistics of Github
only. Github is no measure for GPL.

If you wish to measure how many people eat bread in Austria, you
cannot enter the single McDonalds restaurant and measure it, you need
to measure all the country.

Did you measure https://sourceforge.net/ ?

Did you measure Bitbucket, Gitlab, private, FTP sites, non-GNU
Savannah, and GNU Savannah or Launchpad?

Without proper measurement of statistics, the GPL is not in decline
for me. Because so far nobody have shown real statistics.

If you are speaking only of Github, then I think all this discussion
should be relevant only for Github, it should be maybe discussed on
Github actually.

If you speak of global GPL usage, I have no good statistics, what I
see and observe is that more and more software is released under GPL,
and I am observing it for last 21 years, so my feeling my be
subjective, yet that is my feeling.

Jean