Re: A new practical problem with invariant sections?
On 2/14/06, olive <firstname.lastname@example.org> wrote:
> In every matter, it is virtually impossible to write a rule that can
> mechanically be interpreted to give a suitable result.
It's impossible to cover all aspects of all cases, but obtaining
suitable results is entirely possible.
> The preamble of the GPL is also some king of invariant section:
> it says nothing about the license itself but has only political claims:
I'll agree that invariant sections are similar to licenses.
But I'll not agree that they are equivalent.
> But this shows at least that there can be "sequence of octets" which are
> not the software itself and must be preserved. I claim that the
> invariant sections is just the same: it is not part of the documentation
> (this is required by the GFDL) and there must be preserved.
And we have no problem with maintaining things which are
not the software -- in non-free.
> For the people who don't agree, I would kindly ask them to say if they
> would consider free a license which give you all the freedoms you like
> but must be preserved intact if this license contains a preamble of
> length similar to the invariant section of GFDL? I have ask the question
> in the past but nobody have answered because it was a "facile" argument.
> But if this argument is "facile"; please answer.
It's true that the license associated with a copyrighted
piece of software must be kept intact when distributing that
piece of software. That's a fundamental requirement of
And where someone puts a political rant into an otherwise free
license, we use it as-is.
But the license serves a very specific role -- in free software,
it is what says that the software is free. And we judge the
licenses based on their content.
Invariant sections do not fill this role. They fill other roles.
And the license requirement that they not be removed is
a restriction on modification.
Modifying the DFSG to explicitly allow invariant sections
should be pretty simple, right? The trick would be making the
modification narrow enough that it won't come back to bite
us in a few years.
> The other objections of the GFDL (DRM, etc...) is based on a bogus
> reading of the GFDL.
Whatever "bogus" means in this context...
...and, nevertheless, this is a reading which is could be legally valid.
This restriction should instead say something like "When distributing
the Document, you may not use technical measures to obstruct or
control the rights of other people to read or further copy the copies
you make or distribute."