Savannah rejects a project because it uses GPL

MJ Ray mjr at
Fri Feb 10 18:59:31 UTC 2006

"Alfred M\. Szmidt" <ams at>

> It is usable for free software.  The GFDL has no `advertising clause',

So what would you call the expanded "credit" clause that
seeks to entice legacy publishers to use FDL rather than
a free software licence?

> and has no `encyclopedia' problems,

How could one include parts from the FDL'd Emacs manual in a FDL'd
"Encyclopedia of GNU"? It looks like one must beg FSF's permission,
as relying on a "fair dealing" defence would limit uses.

Rick Moen at TLDP raised this issue in 2004, in
and I'm pretty sure that wasn't the first time it's been raised.

> for some odd reason Wikipedia seems to thrive on the GFDL.

I don't think that's a good example. Even today, many
sites seem to ignore the FDL's terms when modifying
Wikipedia and the Wikipedia FDL story includes questionable
relicensing to remove invariant sections. See near the end of by
Barak Pearlmutter: "In that case the GFDL served to discourage
sharing.  (It nearly prevented it entirely, had remedial measures
of dubious legality not been taken.) [...] The Wikipedia used
the GFDL because it was recommended by the FSF.  They used it
in its natural way.  And then they got burnt."

While arbitrary relicensing is possible if you can claim to be
the copyright holder, like FSF and Wikipedia seem to, it's not
an option open to those who follow in our footsteps.

> There are no restrictions on how you can
> modify a GFDLed manual, the invariant sections are not part of the
> main manual.

So you claim only the manual is freely modifiable (not the invariant
sections), but the manual can't exist without the invariants.  Is
any work which cannot be instantiated in an "all reasonable mods
permitted" way ever free? (This is the "pickle-passing" question
from debian-legal July 2003.)

> The GFDL is perfectly usable for free software, despite your claims
> which you cannot even back up.

Sadly, I can back these up. I called the FDL incorrectly on
day one, had this explained to me in excrutiating detail
(I'm stubborn) and noted my reasons for changing my view.
Because of the sheer volume of material, there are numerous
FAQs that you could read, but I'll try to answer as time
permits if you're unwilling to research it.

> >  > Classifying all licenses as `free' or `not-free' is like saying
> >  > `Intellecutal property', it can mean anything, and it can mean
> >  > nothing.
> >  Actually, I agree with this, including the consequence that your
> >  claim FDL "is a free license" means nothing.
> That doesn't follow.  The GFDL is a _particular_ license, which is
> free.  Classifying _all_ licenses as free or not free is what makes no
> sense.

Now I'm confused. No-one is seeking to classify _all_ licences.

> >  Since the FDL, that just confuses some people who should be
> >  supporting debian and I try to discourage it.
> Considering the problems with Debian and the inclusion of non-free
> software, one shouldn't support it.  Better to support 100% free
> systems like UTUTO-e.

UTUTO-e has included non-free software programs in error (such
as Macromedia Flash and Sun Java - sadly
has vanished) and still includes non-free software manuals.
By design, it's not a 100% free software distribution.
I can't check their current status, as their bug tracker now
requires a username and password.
Debian doesn't include non-free software in the distribution,
promises not to and whenever it happens, that's a serious bug.
The debian bug tracker doesn't require passwords for most use.

Hope that persuades someone to consider debian,
MJ Ray - personal email, see
Work:  Jabber/SIP ask

More information about the Discussion mailing list