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

Re: LPPL, take 2

On Sun, 2003-04-13 at 14:33, Mark Rafn wrote:
> On Sat, 12 Apr 2003, Jeff Licquia wrote:
> >  - 5.a.2.  That's the Clause of Contention, so read it carefully.  I
> > seem to have at least some consensus on it, judging from the feedback so
> > far; its provenance can be seen in this message and the follow-ups:
> I'm close on this one.  "does not identify itself as unmodified in any 
> way" is harder for me to understand than "identifies itself as modified".

Negative is better.  Positive, to me, means "you must write this code,
here" as opposed to "whatever you write, don't pretend to be foo".

> Does "This is LaTeX-format, unmodified" followed a few lines later by
> "this is foo, modified by someguy" qualify?  As written, I'd think this 
> infringes.  

I would say this doesn't (or should not, anyway) infringe. 
Specifically, I would support the notion that a file can self-identify
its own changed status to satisfy 5.a.2.

If you want that to be more clear, how would you word it?

> If the initial LaTeX-format must be modified in order to make certain
> modifications to an LPPL-licensed module, it's hard for me to see this as
> a free license.

Agreed here.

> >  - 5b.  Mark, you were nervous about this, but I don't see an
> > alternative or clarification in the discussion.  Are you satisfied, or
> > is there still some work to do?
> I think my objection to 5b boils down to the fact that it doesn't
> distinguish between API strings and user-copyright strings.  As long as
> the package contains no must-modify strings which are part of the
> container's API, I don't object.  I'd strongly prefer this were clarified
> in the license.

With the language in 5.a.2, is 5b now irrelevant?  Changed files are
already required not to identify themselves to the user as unchanged,
which would seem to cover the case where "identification strings" were
displayed to the user.

We might want to keep 5b in a lesser form: requiring notice in the file
itself of changes, but without any requirement to display such a thing. 
This would be similar to the GPL's clause 2a.

What do we think of this?
Jeff Licquia <licquia@debian.org>

Reply to: