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

Re: Draft new DFSG

On Tue, 24 Nov 1998, Marcus Brinkmann wrote:

> On Mon, Nov 23, 1998 at 04:21:51PM -0500, Dale Scheetz wrote:
> > On Mon, 23 Nov 1998, Ian Jackson wrote:
> > 
> > > As I've said before, most recently in a posting to -private, and
> > > before that in various fora, I think the DFSG has some serious
> > > problems due to loose wording.  I also strongly dislike the patch
> > > clause.
> > 
> > I have heard this from you before, and I don't think you understand just
> > what is at stake here. Let me see if I can explain.
> [explanation snipped --- I really just seek an entry point for my NSHO :)]
> > We should not declare a license non-free which only requires that the
> > original source be propogated unchanged. I would argue that this provides
> > more freedom to the "end user" since it should then be a simple matter for
> > the "current owner" of the sofware to refert to the original author's
> > work.
> What I dislike about "patch only" distribution is twofold:

The above constraint does not require "patch only" distribution. You could
just as well provide a complete modified source beside the original work.
As this is space consumptive, the patch method is a more functional
process, but not required by the initial conditions.

> * Companies like to enforce different copyrights on patches than on the
>   original work (see NPL, see QPL, the latter is even worse, because you
>   give up almost all rights on your patches. This is bad, because the
>   patches are not any longer protected like GPL software or the Qt source if
>   that matters.).

This is protected by other clauses in the license as described in the
DFSG. The GPL "protections" are sometimes viewed as "infections" because
of the way the GPL propogates into derived works. A license that requires
unmodified source, can still have these protections.

> * In the long run, you end up with a long upstream source and even more
>   patches. This is at least uncomfortable (I speak of cases in which the
>   source is not maintained by the original copyright holder anymor).
I see this argument as equivalent to "why should I be forced to supply
source?", it is not needed by most users, it takes up space on the
distribution medium (or wherever you need to store it so that it is

You provide source for reasons of freedom, requiring unmodified source
provides another degree of freedom by requiring that the original author's
work be propogated unchanged. This is an additional protection from
monopoly by the agency making changes.

Consider: distribution A creates modifications to a particular package to
support its own goals. This distribution obtains dominance in the market
place, and soon becomes the only source for that particular package. Folks
making a new distribution prefer to use the original design, as it fits
their purposes better, but the only source available is from distribution
A, and there is no original code available. (Note that this is
particularly important for code where the author is no longer
available...like dead)

Particularly in a typesetting program where uniformity is an asset, these
conditions of "no change" are desirable and valuable. To decide that such
code is "non-free" ignores the value and freedom delivered by such

Waiting is,

_-_-_-_-_-   Author of "The Debian Linux User's Guide"  _-_-_-_-_-_-

aka   Dale Scheetz                   Phone:   1 (850) 656-9769
      Flexible Software              11000 McCrackin Road
      e-mail:  dwarf@polaris.net     Tallahassee, FL  32308

_-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_-

Reply to: