firefox, iceweasel, burningdog, icecat, ...

Yavor Doganov yavor at gnu.org
Mon Nov 3 20:04:23 UTC 2008


Sam Liddicott wrote:
> 
> > No, because it was the only way to go forward, 
> ? You just admitted that it was probably possible.

In theory anything is possible.  Maybe it was possible not to use a
non-free compiler too in the old days, perhaps at the cost of dealying
the development of the OS for years, or going into a completely wrong
direction, discovering that much later.

Of course, as you say, it is an assessment of pros and cons based on
personal values (of the one making the decision).

It is one thing to modify GNU make to build itself without a `make'
program at all.  It is different to compile a compiler without an
existing compiler, at least at that time.  It is quite a bit different
to _develop_ a compiler without being able to compile and test it in
phases.

> > It was not an easy decision.  
> if it was impossible it would have been an easy decision.

Right, although any decision that is in conflict with the goal cannot
be "easy", no matter the circumstances.

> > Fortunately, nobody has to face such dillema today.
> >   
> This is the point we are discussing.
> I argue that for many, there are dilemmas of the same abstract form
> today.

I understand your point, and here we disagree: there is no reason for
a user or developer or $insert_role_here to make such dillemas today.
There are certain sacrifices you must do if you want to be a free
computer user, and that will be the case for some years to come.  The
magnitude of these compromises is dimminishing very quickly.

> You seem to suggest that they don't count BECAUSE those facing the
> dilemma's are users, not developers of free software.

No.  I am a user myself, not a developer, so the most natural point of
view for me is the user's.

I think that here you are making a fundamentally wrong comparison: the
compromise back then was necessary for the development of the
foundation blocks of the GNU system (which is, more or less, what
distros install today as "minimal" or "standard" system).  When you're
missing one of those major blocks (such as libc, kernel, compiler,
shell, binutils, build system, core utilites, etc.), you're in
trouble.

Once you have those, you can build the rest without relying on
non-free software at all.

> I do suggest that if gnu/linux and iceweasel is used with non-free
> flash that there are more free software users than if ms windows and
> explorer are used with non-free flash.

If you weigh on the quanitity, that is true.  But it's an illusion.
More free software users does not lead to our goal being reached
faster or earlier.

> Another aim of free software is the liberation; which cannot be done
> merely by writing software, it requires evangelists;

This is not another aim, it is the only aim.

If people do not recognize non-free software as a social problem, we
have failed in our mission.

> it means to win the understanding of the ignorant, it requires
> teaching, and this is quite a different problem,

Here is the main point of the whole disagreement:
It appears that you think that if you manage to persuade a user or N
users to switch to free software (enitirely free or not, it does not
matter), that is a net win for the free software cause.

I say that this is nothing.  Really.  Free software will become more
and more common, for various reasons -- one of them being that it is
becoming technically better, and this process is fast.  We don't have
to do anything special about that; no efforts are required.

Users who switch to free software for non-philosophical reasons (like
the company I work for, or my uncle) cannot possibly defend the cause
or keep up the community, because they're ready to trade away the
freedoms they have (because, as you say, they don't realize them and
don't value them).

So what's the point in getting more users like these?  The Open Source
campaign is doing already enough in this regard.  It is of course not
a bad thing at all -- but it's not our goal as activists, and more
importantly, does not help the mission.

> The first commonality is the value-judgement, or rationalization of
> the compromise which occurs in both cases; strangely you both denied
> and admitted that such a compromise occured in the early days of
> developing the tools.

Refer to my explanation above wrt "must have foundation" and "the rest
of the system".  That is how the subsequent packages were developed
(e.g. GNUstep, GNOME, etc.) -- nobody had really faced this problem.
(I admit the presentation of my opinion was not clear, and maybe still
isn't.)

> The second commonality is that the compromise in both cases is over
> whether or not making the compromise will advance the aims of the
> free software movement.

Yes.  While this is obvious in the GCC case, I still can't see even a
faint sign that it is in the example of Firefox + Flash.  

> it also affects the introduction of others to free software.

Such introduction does not help much (or should I say "at all") if
these users do not embrace the values of the free software movement,
and are not ready to fight for them.  Suggesting them to treat such
non-free software as acceptable only undermines the goal: they will be
introduced to free software, and they'll be using it (mostly), but
they won't value what they have and what they may lose.  Such users
are at constant threat of being enslaved again, perhaps even without
noticing as they cannot spot the danger.

> Lets not dispute about the language, but from my frame this is an
> admission that you don't appreciate someone elses requirements.

That's the namely the point: if a mere usage requirement leads to
neglecting ethical principles, then these principles are not so strong
or beleived in the value system of the person making the decision.
Rather than weaking the principles, we'd better convince people that
they're worth as they are.

It is strictly personal, and my opinion is that helping the spread and
adoption of free software is a very small part of the job, as "thanks"
to the open source guys the values of the free software movement do
not spread at the same rate as the software.

> My young children are an example. Right now captivity is more
> attractive to them. I do not want to teach them to hate freedom. So
> I make a compromise which I feel will promote free software in their
> life. 

I do not refuse mine to visit any site they wish, I just explain them
why they can't view certain content, and why I cannot install the
software required to view it.  I don't think they understand yet.

> Perhaps we can agree that users who don't yet understand freedom and
> liberty are not enemies of the free software movement?

Of course, although without generalizing so much.  Some of our
outright enemies do not understand software freedom too.

> Perhaps we can agree that those who help such people make
> compromises that speed their migration to free software are also not
> enemies of the free software movement.

Those seem to be the open source people, mostly.  I don't treat them
as enemies, I just think that by rapidly "helping" people this way and
speeding the migration, they are making our job harder.  It was
substantially easier a decade ago, when people already had to make
conscious choice and a convenience sacrifice to use GNU -- and it was
easier to persuade them, as they approached the issue with a clear
mind, juding very often solely based on the ethical values.

> http://torvalds-family.blogspot.com/2008/11/black-and-white.html

Nothing new on the horizon; people insisting on software freedom are
treated as extravagant creatures since at least 1998, if not before.

> but what is the FSF going to do about people who aren't there yet?

It does lots of things in that regard; in fact almost all of its
activities are diverted in that direction.




More information about the Discussion mailing list