Re: Concerning uniliateral decision making with respect to the coreutils package

Simon Josefsson <[email protected]> Mon, 13 Jul 2026 22:30:44 +0200
Newsgroups	gmane.linux.debian.devel.general
Message-ID	<[email protected]>
Russ Allbery <[email protected]> writes:

> Simon Josefsson <[email protected]> writes:
>
>> To me, I find free software more of a philosophical, political and
>> social choice than technical.
>
>> I use proprietary devices with proprietary software (e.g., macOS or iOS)
>> and by almost any metric any random person could think of, they are
>> "better".
>
> I think macOS is vastly inferior to Debian in many respects, so I would
> not concede that, but that's beside the point. (iOS, sure, the free
> software situation for mobile phones isn't great.)
>
>> Thus, to me the reason to prefer (strongly) copyleft software is not
>> about getting "better" software.
>
>> I think that if we don't have, and promote, strongly copyleft software,
>> we risk end up being subjugated by those who want to control my software
>> usage.  Through tivoization or some more modern method.
>
> Right, I understand all this. So does everyone else. I would be surprised
> if there is anyone reading debian-devel who has not heard these arguments
> more times than they can count.
>
> You are not wrong to feel this way. I am not disagreeing with this
> argument. I am pointing out that regardless of whether *you* care about
> having better software, you are trying to convince *other people* to use
> copyleft software, and many of those people *do* care about having better
> software.

That wasn't my goal here.  I think people should be allowed to use any
software they want, including free or proprietary software.  It is
better if they are able to make an informed choice, but that's hard to
demand.

My point earlier in the thread was that by replacing (strongly) copyleft
software by non-copyleft software (and I include weak-copyleft in the
non-copyleft category) in Debian there is a risk the technical
excellence of Debian will be used to subjugate users, and that
eventually this could erode the foundation for why Debian is technically
excellent in the first place.

Having (strongly) copyleft software in Debian is one gate-keeper against
Ubuntu wrapping it up in a commercial offering and not offering source
downloads.  Canonical migrating applications to snaps as a "better"
software distribution method is an indicator they want something that
the existing methods does not give them.

I think a Debian core without any remaining (enforcable strongly)
copyleft components a not unlikely scenario in the next ~10 years.  The
move towards that has been happening for many years already, using
various arguments including changing to "better software".  bash, gawk,
wget, libgcrypt, info, etc.  The market forces are there to promote
this.

Maybe it will take 20 years rather than 10, but I feel this is
inevitable.  I don't see it as evil but merely a logical progression.  I
mostly enjoy this too: older Debian contained a lot of unsustainable
things.  bash and gawk are slow or complex compared to dash and mawk.
Seqoia PGP is better engineered than GnuPG.  Clang has a lot of features
lacking in GCC.  I struggle to find good things to say about OpenSSL
compared to libgcrypt or nettle, but I suppose a large user base
improves security.  Those are all good arguments.  Still I find the
trend is problematic from a user freedom perspective, and
counter-productive in the long term.

> You can think that they shouldn't care and preach yet another round of the
> copyleft sermon, but when you do that to an audience who has heard that
> sermon literally thousands of times, you can't expect this to accomplish
> very much.
>
> Originally, one of the big appeals of GNU software is that it was *good
> software*, which was also copyleft. In fact, one of the original arguments
> for copyleft is that it would create a virtuous circle of development by
> forcing new work to also be free software. That was a quality argument. A
> lot of people started using that software because it was high quality and
> then got curious about how that was done and became believers in the
> underlying philosophical structure as well. GNU coreutils continues
> following this model to this day, and I think quite successfully.
>
> There is no substitute for writing good software. Telling people to use
> worse software because it's copyleft will convince some people up to a
> certain point, but if the copyleft software is clearly inferior, well, the
> number of people who *only* care about using specifically copyleft
> software is, at least in my opinion, not that large.
>
> You can keep trying to convince those people they're wrong to have that
> attitude, I guess, but I don't think it's going to work. Certainly not by
> just repeating the same arguments they've heard many times before.

That's the same arguments we use to convince people to use Debian
instead of Windows or macOS, isn't it?

Debian is "better" because of X, DFSG, Y, DSG, Z.

Many people disagree, and continue use macOS or Windows, and find the
arguments for using Debian non-convincing or not applicable to where
they are.  I can't blame them.

I don't see that situation as a reason for not continuing to offer
Debian as an alternative, do you?

In the same way I think we could continue to offer (strongly) copyleft
software as an alternative.

Some minority will find Debian useful for them, some (probably somewhat
overlapping) minority will find copyleft software useful.

I think optimizing for popularity is a common fallacy in these
discussions.

/Simon
signature.asc (application/pgp-signature, 1.2 KB)