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

Re: Alternatives to the Affero General Public License

On Wed, 22 Jun 2005 02:52:55 +0000, "Brian M. Carlson"
<sandals@crustytoothpaste.ath.cx> said:
> On Tue, 2005-06-21 at 19:05 -0700, Gregor Richards wrote:
> > Because the AGPL has some implementation issues that make it possibly
> > incompatible with the DFSG, I've been trying to find an alternative that
> > would still protect source-code redistribution on line.  Basically, I'm
> > trying to write a special exception to the GNU GPL that would add this
> > without some of the practical problems, and possibly with DFSG and OSI
> > compliance.  I'm not entirely clear on to what extent point 4 (Integrity
> > of The Author's Source Code) applies.  Clearly, the AGPL creates an
> > "invariant section" like the GFDL, which doesn't work.  My proposed
> > change works more like clause 2(c) of the GNU GPL.  There's no exact
> > code that needs to be kept, but a certain functionality does.  I don't
> > think this contradicts anything in the DFSG, but I'm no expert, and
> > would like your opinions.
> > 
> > 
> > 
> > Here's my proposition so far:
> > 
> > 
> > As an exception to the GNU General Public License, any user who wishes
> > to distribute modified copies of <program> (the Program) is also
> > required to abide by this additional rule:
> This is an additional restriction, forbidden by section 6.  Section 6
> states, in part:
>   You may not impose any further restrictions on the recipients'
>   exercise of the rights granted herein.
> This results in having an inconsistent license, and is consequently not
> distributable at all, except by the copyright holder.
> Also, the word "insure" should be "ensure", on the fourth line of the
> first starred item.  The word "insure" means "to secure against loss".
> The word "ensure" means "to make sure" or "to be certain".
> As a side note, what you are trying to do is not compliant with the
> DFSG.  We have rejected software that requires that people send copies
> or otherwise make copies available[0] as compatible with the DFSG.
> Whether it is compliant with the OSI is not a discussion for this list.
> [0] See http://wiki.debian.net/?DissidentTest
> -- 
> ($_,$a)=split/\t/,join'',map{unpack'u',$_}<DATA>;eval$a;print;__DATA__
> M961H<F$@8FAM;"!U<F%O<G-U(#QU<F%O<G-U0&=D:75M<&UC8VUL=G)U;6LN
> 5:75Q96AT9V1Y>F%L=G-P;6IX9BP)

In response section 6:
(So that I can reference, the full section:)
6.  Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions. You may not impose any further restrictions
on the recipients' exercise of the rights granted herein. You are not
responsible for enforcing compliance by third parties to this License.
  It seems to me that the license from the original licensor would
  include this new term/condition, as that is how (s)he licensed it. 
  Perhaps a rephrase as "an additional term" or something that made it
  obvious that this was a part of the license of the program and hence
  is not an additional restriction when redistributed?  I of course
  can't make an entirely new license based on the GNU GPL without FSF's
  permission, so is there any way that a term could be added at all?

In response to {i,e}nsure:
Of course you are correct, my current draft now has this changed.

In response to the dissident problem:
I don't see how this hinders said dissident at all.  If said dissident
has to send the entire source, (s)he as already made it available
through some computer network.  If said dissident made it available on a
public computer network, they have already incriminated themselves - if
said dissident made it available on a private computer network, the
distribution of source does not need to extend beyond that network.  If
said dissident simply made changes and had it on no computer network,
(s)he could distribute it as described in that wiki page.  It seems to
me that by committing to send source code via the same interface, users
are not committing to any more than they already have by having the
program on that interface.  Perhaps I don't understand the application
of this rule?

 - Gregor Richards
  Gregor Richards

http://www.fastmail.fm - Or how I learned to stop worrying and
                          love email again

Reply to: