Re: Possible draft non-free firmware option with SC change
Simon Josefsson <simon@josefsson.org> writes:
> Thanks -- this helps me understand the two principles at play here:
> 1) having a free Debian
> 2) having a Debian that works on as much hardware as possible
This summary is moving in the right direction! But your phrasing of 2)
isn't the principle that I personally hold; it's a consequence of that
principle. I would rather go one level deeper and phrase it something
more like this:
2) having a Debian that is useful for and supports the needs of its users
In other words, the principle for me isn't about hardware or any other
similar specific problem. It's about making Debian usable in the general
sense. For me, this is an ideological principle in support of free
software: I believe the only way to extend the spread of free software and
make it an ideological force in the computing world is to make software
people can actually use, while making it as free as possible without
making it useless to them.
When I first got deeply interested in free software in the late 1990s, I
looked around and saw two basic mindsets towards free software. I'd
classify those as the FSF approach and the Debian approach. The FSF
decided to go down the route of ideological purism: they made the absolute
minimum number of compromises possible and then shed them as soon as
possible. Debian instead took the route of practicality and tried to make
the operating system usable and flexible, recognizing that sometimes for
some people that would include non-free software. That upset the FSF
quite a bit; they considered (and I believe consider) Debian to not
"really" be a free software project because of this stance.
My opinion then, and my opinion now, is that Debian has the better of that
argument. Debian's approach is simply more effective *at promoting free
software*. As a result (not only of that stance, but largely I think
because of that stance) the FSF's attempts at producing operating systems
have been hobbyist experiments and ideological statements that almost no
one uses. Meanwhile, Debian has become the foundation of numerous major
free Linux distributions.
> For me, principle 1) is more important than 2). For you and Steve, if I
> may put words in your mouth, principle 2) is more important than 1).
Ah, no, I have explained this poorly. This is not at all true for me, and
I suspect also not true for Steve.
For me, principle 1) is *equally important* than principle 2), and my
disagreement with you is that I feel like you're discarding principle 2)
rather than giving it equal weight.
> I don't think the principle in 2) is well supported by Debian
> documentation.
I believe that you're missing point 4 of the Social Contract.
We will be guided by the needs of our users and the free software
community. We will place their interests first in our priorities. We
will support the needs of our users for operation in many different
kinds of computing environments.
The purpose of the Debian Project is absolutely to create a free software
distribution. We are not Apple; the point is to build on top of free
software. If I really considered 2) much more important than 1), I'd be
in favor of rolling non-free into main, including non-free drivers, and so
forth. I am not.
But Debian, very early on, decided to navigate the tension between those
two equally-held principles by taking the route of making it usable
*first* and then as free as possible. The guiding principle where we have
options about how to do something where both can work for the user is free
software; that's the point of this endeavor. But we don't tell users that
their hardware is useless and they need to buy new hardware in order to
maintain free software purity. We meet them where they are, and then help
them make their system as free as possible.
--
Russ Allbery (rra@debian.org) <https://www.eyrie.org/~eagle/>
Reply to: