[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Why is Fedora not a Free GNU/Linux distributions?



On Jul 24, 2008, Les Mikesell <lesmikesell gmail com> wrote:

> Alexandre Oliva wrote:

> it has always been immoral to demand that others give up their rights.

Taking away legitimate rights, yes, that would be immoral.  But taking
away Immoral rights, that serve the purpose of exerting power over
others and taking away their rights and freedoms, no, nothing immoral
or wrong about that.

>> Other copyleft and non-copyleft licenses followed that were designed
>> specifically to be incompatible with the GPL, to prevent code sharing.

> BSD license whose intent is obviously to permit code sharing but not
> take away anyone's rights predates the GPL and certainly GPLv2.

The BSD licenses are not copyleft licenses.  They are permissive Free
Software licenses.  One of them is GPL-compatible, the other isn't.

> The GPL was intentionally incompatible because it's only purpose is
> to force you to choose between giving up your rights or sharing
> code.

You can't give up a right you never had.  The GPL (or any Free
Software license) doesn't (and can't) take away any right you had.

Its purpose is to ensure that all recipients have the four freedoms
respected in all programs based on the program licensed under it.
Permitting the program to be distributed under whatever terms you like
would grant you the power to not respect others' freedoms.  I'd much
rather you didn't have that power, and if you are a well-meaning
person, respectful of others, you shouldn't mind that you don't.

>> Can't fault the GPL for that, can you?

> Of course I do.  There's never a reason for anyone to give up their
> rights, and no need to withhold  the ability to share code.

If everyone was respectful of others, there wouldn't be any such
reason, indeed.  There wouldn't be reasons for laws either.
Unfortunately, some people are too selfish, don't think of others, and
need society to put some limits on them so that they get stronger
reasons to respect others' rights.  That's the reasoning behind
copyleft.  It has its drawbacks and inconveniences, no doubt, but
we're much better off when someone has to decide between respecting
others' freedoms or incurring the costs of duplicating code in order
to not respect them: at least some of them will then decide to respect
others, and, for those who decide not to, they wouldn't have respected
them in the first place, so we lose nothing from them.

So the only actual loss is from incompatibility between Free Software
licenses.  As I explained on the thread from one month ago on
fedora-devel, exceptions are often possible (see 10. in GPLv2), and if
they're too hard to obtain, it's because of legal maintainability of
the projects, or because the licensors aren't really interested in
respecting your freedoms (this unfortunately happens), or in granting
you power to make it easier for you to erode others' freedoms (this
fortunately happens).

>> If you think that's
>> unacceptable, why did you accept the less permissive or artificially
>> incompatible license that got you into this situation?

> In general the terms I'm speaking of are more permissive than the GPL

If they were strictly more permissive on all accounts, then they would
be compatible.  The original BSD license, for example, with the
advertisement clause, didn't create a major problem as long as only
UCB made that demand.  But as you combined code from multiple sources
with multiple such demands, it would soon have become a major hassle
to abide by all those advertisement clauses, to the point of making it
prohibitive.  Now, the modified BSD license, that came up after years
of negotiations between UCB, the FSF and others, lifts this burden and
thus becomes compatible with the GPL.

> The point is that the work-as-a-whole clause is an immoral
> restriction.

Granting permissions over the work as a whole is immoral?

> The thing that's not acceptable is forcing a decision to give up your
> rights or to not be able to share the code.

The only situation in which you're not able to share the code is when
it's part of a work that is based on someone else's work.  You don't
have a right to share that without permission from this someone else.
If you pretend that you do, you're misleading and confusing yourself
and others.

> Depending on your legal system there may or may not be a difference
> between a license and a contract, but having agreed to:

>   "b) You must cause any work that you distribute or publish, that in
>     whole or in part contains or is derived from the Program or any
>     part thereof, to be licensed as a whole at no charge to all third
>     parties under the terms of this License."

Ok, that's 2b.  [I realize I responded before as if you were referring
to 3b, sorry.]  Now, please realize that this is not to be taken in
isolation.  This is a condition on one of the various grants of
permission for you to copy and distribute the program.

  2. You may modify your copy or copies of the Program or any portion
  of it, thus forming a work based on the Program, and copy and
  distribute such modifications or work under the terms of Section 1
  above, provided that you also meet all of these conditions:

There are other permissions that enable you to copy and distribute the
program, such as 1. and 3.  As long as you operate under any one of
them, you're covered.  And, if you have any other permissions, granted
by other licenses or arrangements, you can exercise them too.  The GPL
doesn't prevent you from doing so.  What you quoted above is just
saying that, if you abide by those conditions, then the GPL grants you
these permissions.

It doesn't say that you have to agree to never do different from that.
Say, you can still breathe after you accept the GPL, even though the
GPL doesn't say you can.  And if it did say something like:

  2+i) You may modify your copy or copie of the Program or any portion
  of it, thus forming a work based on the Program, and copy and
  distribute such modifications or work under whatever terms you like,
  provide that you meet this condition:

  a+i) you stop breathing immediately.

(nevermind that such a condition is unenforceable, let's pretend it
 isn't, for the sake of the argument)

Now, if you accept such an hypothetical IAGPL (Imaginary Asphyxiating
GPL ;-), would you be agreeing to never breathe again?  

No, you wouldn't.  There's nothing in it that forces you to stop
breathing the moment you accept the license.

At this point, let's assume hypothetical Les Mikesell asks:

] But what if I distribute the program?  Don't I have to stop
] breathing immediately?

No.  If you didn't stop breathing, and I (let's assume I'm the
copyright holder) sue you for copyright infringement because you
evidently didn't stop breathing, your lawyer will tell the court:
"See, your honor, my client chose to distribute the program under
clause 2, not 2+i, and he complied with the conditions set forth in
it."  The judge would then verify that you have indeed complied with
those terms, and that therefore you had indeed permission to
distribute the work the way you did.  I'd be laughed out of the court,
regardless of whatever clause 2+i said.


Now, going back to the real GPL, version 2, that you quoted.  1.,
2. and 3. are separate, mostly independent permissions (except whey
they explicitly refer to each other).  If you do what they permit,
you're covered, regardless of the conditions set forth in any other of
the permissions.  (Which is not to say that any other clauses don't
apply.)

So, if you have other permissions, you can enjoy them, regardless of
the conditions set forth in the permissions you don't want to enjoy.

> consider all of the possibilities you have given up the right to
> create and share.

Easy.  None whatsoever.

> And here's a more pragmatic take on the issue.
> http://www.oreillynet.com/pub/a/policy/2001/12/12/transition.html?page=1

You'll note the prevalent use of open source, and the clear motivation
on "perhaps getting more contributions from commercial corporations"
[who don't want to respect others' freedoms] rather than "respecting
others' freedoms".

No wonder they may get to different conclusions than those whose
values they don't share.

Some key points to their assessment is:

  those who learn to cooperate will benefit the most. We believe that
  this cooperation should be based on mutually acceptable terms, and
  that the best basis for cooperation is voluntary contributions.

  There are plenty of other examples where commercial corporations
  have contributed voluntarily (with or without copyleft)

  Although we believe that you are more likely to receive
  contributions from commercial corporations if you abandon copyleft,
  we can make no guarantees that this is going to happen.

  Based on our experience, we advise open-source developers to use the
  least amount of copyleft necessary.

I.e., anecdotal evidence for an unsupportable belief, most likely
already held before, that they'd get more commercial contributions if
they "permitted commercial exploitation", while fear of commercial
exploitation was their primary motivator (misguided as it was) for the
choice of a strong copyleft license in the first place.

Do I need to point out the contradiction here, or is it obvious
enough?

-- 
Alexandre Oliva         http://www.lsd.ic.unicamp.br/~oliva/
Free Software Evangelist  oliva {lsd ic unicamp br, gnu.org}
FSFLA Board Member       ¡Sé Libre! => http://www.fsfla.org/
Red Hat Compiler Engineer   aoliva {redhat com, gcc.gnu.org}


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]