On Sun, Mar 07, 2004 at 09:45:43PM -0800, Thomas Bushnell, BSG wrote:
> Anthony Towns <aj@azure.humbug.org.au> writes:
> > I think [foo] but the mere possibility of [bar]
> > isn't a problem even if we decided [baz].
> So your position is that we should have non-free for as long as there
> is any doubt whatsoever if there will ever be a package to place in it?
No, my position is that it'd make sense to remove non-free when there
are only a handful of packages to be kept there, because it's likely
none of them would be particularly interesting and the effort of keeping
non-free in the archive wouldn't be justified. [foo]
But even if Debian adopts a more extreme position, namely that non-free
shouldn't be removed while there's a prospect of non-free software being
packaged and useful for our users [baz], that doesn't cause any problems
because as long as the trend remains that non-free software is becoming
less and less necessary, we'll eventually reach a day where that condition
is satisfied [!bar].
> And I don't know how we could ever have that confidence, unless the
> copyright laws get changed, because someone could always write
> something and make it non-free but distributable.
First, I'd hope that over the long term we do change our copyright
and patent laws. I've got no problems with keeping non-free until that
happens.
Second, the way you ensure these things in a free society is by ensuring
that people realise it's in their best interests to be writing free
software: by demonstrating that it can be more profitable, less costly,
more effective, and more flexible. There are a bunch of cases where
we can't demonstrate that yet. I believe if we give free software a
reasonable chunk more time, we will be able to demonstrate it, and I've
got no problem with keeping non-free until we have done so.
> Perhaps I've misunderstood. Is there some minimal number of packages
> such that if we have only that small number, we can disregard them and
> close down non-free, in your opinion?
No, not particularly. The cutoff is when the administrative burden of
worrying about non-free becomes more than it's worth to its users; I'd
suspect that'll come when there's but a handful of packages there, but it
might come sooner (if there are a couple of dozen packages that are all
pretty pointless), or it might come later (if we have one or two packages
that are really important to some users that are really hard to replace).
> > That's the system we've already got -- people don't like maintaining
> > non-free software, so when there really is some free software that fills
> > the same niche, it gets dropped by the maintainer. If you'd like to do QA
> > work making sure that happens more promptly than it does atm, please do.
> In practice, this is not true. Often there is a different maintainer,
> who continues to maintain it because he likes it, completely
> independent of whether there is a free alternative.
If he still likes it, then it does some things better than the free
alternative.
> Netscape did not
> get dropped because free web browsers became available;
Netscape, eg, worked with various plugins better than mozilla did.
Cheers,
aj
--
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.
Linux.conf.au 2004 -- Because we could.
http://conf.linux.org.au/ -- Jan 12-17, 2004
Attachment:
signature.asc
Description: Digital signature