On Mon, Jun 24, 2013 at 07:51:51AM -0400, Bradley M. Kuhn wrote:
Richard Fontana wrote:
> I've been trying to do something I believe I talked about before,
> re-implement copyleft-next as a 'file-based' copyleft that is less
> minimalist than MPL.
That's unfortunate. I don't think the world needs more weak copyleft
licenses.
Good to see you again, bkuhn!
It has enough already. Shouldn't you rename the project to
weak-copyleft-next or lesser-copyleft-next then?
A realization I feel I may be coming to is that the distinction
between 'strong' and 'weak' is no longer as significant as it may have
seemed in the early 1990s when GPLv2 and LGPLv2 were released.
I do recognize the force of the criticism that there are already a lot
of copyleft licenses that appear to have been conceptualized as
'weak'.
However, I see copyleft-next as potentially something *beyond* the
strong vs. weak paradigm. I may have more to say about this.
A bit more constructively: file-based copyleft is always easy to
circumvent.
As regards MPL-style file-based copyleft, this is well-known and
indeed apparently justified (at least by the time of MPL 2.0) as a
matter of policy. The GPL, as conceptualized by RMS, seems to be meant
to be *possible* to circumvent (the GNU licenses FAQ is known to be
used [perhaps naively] by some lawyers as a roadmap for counseling on
how to avoid the effects of strong copyleft) but *difficult*. I
continue to believe 'making copyleft circumvention difficult' is an
important goal (for some copyleft licenses).
But I am not yet willing to concede that a file-based copyleft could
not be materially less easy to circumvent than MPL 2.0.
I think the beauty of strong copyleft is somewhat centered around
the
round-peg/square-hole scenario we get when we apply the derivative and
collective works doctrines to software. I don't think we can get that
effect by implementing detailed rules in the license.
Aren't you basically saying that strong copyleft is by design supposed
to be difficult? I feel strongly that copyleft rules should not be too
difficult for well-meaning people to interpret and apply.
I actually think you could implement detailed rules in a license and
still get roughly the kind of strong copyleft you would probably like
to see, but it would be a *very* long license -- yet shorter than any
number of negotiated commercial technology license agreements. (One
might argue that GPLv3, nearly twice as long as GPLv2, is in part an
effort to make the GPL a bit more 'rule-based' [I may be stretching
the rule vs. standard distinction there].)
- RF