[linux-elitists] The GPLV3 Position... (and a suggestion to Jon Corbet)

glen martin glenm@locutory.org
Wed Sep 27 06:51:28 PDT 2006

Dave Crossland wrote:
> On 26/09/06, glen martin <glenm@locutory.org> wrote:
>> But I can redistribute it, modulo my
>> question about p9 above, under some new license of FSF's choice, which
>> may bear little or much similarity to the GPL v.2 the contributors
>> originally used.
> It bears much similarity, thus it is the next version of the GPL.

GPL v3 is not "the next GPL" because it bears much similarity to v2, but
only because the FSF call it so.  Otherwise LGPL was the next GPL.

My thoughts around "much or little similarity" were intended to be a
little more generic, and should probably have been worded differently to
make that point better.

As a license is revised, any similarity or otherwise in goal and effect
are the responsibility of the author - contributors by and large don't
have a say. So if Zaphod Beeblebrox were to rewrite ZSL (Zaphod's
software license) v1 to make a very different v2, that's his prerogative.

Others don't have to use the new license. That's theirs.

But if Project ZZ9-Plural-Z-Alpha uses ZSL v1, me, a user, hacker (in
the original sense), and all-around good guy, could use ASL v2 on my
redistribution. Without permission of contributors to that project,
though Zaphod would no doubt be charmed.

All because of a name. Different license. Same name.

I can't take any other different license and substitute it that way.

The term "ZSL" (or "GPL") is a brand, and just like the term "Websphere"
which can be applied to every product from appservers to tennis balls,
it can be applied to whatever the owner of the brand likes. Even a
completely different license.

> In a very general sense, the controversy about GPLv3 seems to come
> from a new generation of software proprietors who evolved under the
> selection pressure of GPLv2 (web app proprietors, special-purpose
> computer proprietors) and who are are getting annoyed by the
> reassertion of its original aims.

I disagree with your assertion that to argue is to be annoyed at a
reassertion of original aims. Such a statement might be reasonable if
the changes were clear bug fixes (eg. "this wording has failed when
challenged in court" or such). But "original aims" has not been established.

I think correspondence to the original aims is arguable. Indeed, it is
being argued by a number of kernel contributors that the new (draft)
license doesn't meet their needs. I believe the argument includes a
notion that the original aims were said to be about rights to make use
of software, not about hardware (tivoisation). This seems a difference
in aims as expressed by text.

When I've read licenses to select one for a new  project, for me it has
often come down to single clauses. This license B seems to have been
written to address a percieved problem with that license A, and the
difference comes down to a few significant words here or there. 

Independent of what may be in the mind of the author, the text as a
whole defines a binary condition: the license meets my needs, or it doesn't.

Since you brought up Microsoft: if MS were to say their new purportedly
open source license is intended to accomplish a particular goal, would
that claim matter to you? You'd read the license, and make decisions
based on what you found there. Not elsewhere.

Expressions of "aim" aren't worth much, IMO.  In a very practical way,
the text, with its various terms, benefits, obligations and so forth, is
the only useful expression of the aim.


More information about the linux-elitists mailing list