Rafael Laboissiere writes:
> * Maurice LeBrun <mjl@...> [2003-08-16 20:28]:
> > I agree a warranty clause is needed. I am worried the previous paragraphs
> > are a bit too restrive tho. In particular, I have personally been in a
> > situation in the past which is a little difficult to reconcile with
> > stricter (than we have now) conditions on republication. Let's say a
> > researcher at a govt lab puts considerable effort into making a plplot/Tk
> > interface for a simulation / analysis tool, and wants to issue the plplot
> > manual as an internal report subject to internal publishing guidelines,
> > language translations, etc. It may even be a part or appendix of a much
> > larger work. Under these conditions, the issue of distribution of the
> > entire work becomes complicated.
> You are right, but I think that this is more a virtue of the license than a
> problem, since it avoids hijacking of the manual. I do not believe that
> this situation would happen in the future, but since we want the manual to
> stay free (in the Open Source definition) we better take the precautions.
In a /particular/ open source definition, you mean. There are many open
source definitions, and many that are less restrictive than GPL and it's ilk.
For software projects I personally work on, I don't particularly even like the
GPL, only the LGPL is tolerable to me. My main concern is to not overly
burden the code / docs with restrictions. *Especially* the docs, since
there's no point to it.
As for "hijacking", that's a loaded term and is not applicable. No one can
seize control of the manual. At "worst" they could customize it and take
it commercial. /Just/ the docs. LOL. Good luck to them.
> > I also don't see how someone sending me a page full of text in Chinese and
> > asking me if this translation is correct is going to be worth my time.
> I know a couple of trusted Chinese Debian developers that could do this for
> us quickly (remember that only the translation of the permission notice
> requires approval). This is valid for most of the languages in the world,
> in which the manual could potentially be translated. Also, I think that
> people at FSF have already coped with the issue.
OK, so Lithuanian. Swahili. Mongolian. I don't care, I don't want to be
bothered by translation issues, period.
> > I'm well aware of the dangers of having no limits on the freedom to modify
> > and redistribute. However that mostly applies to programs; without the
> > program the documentation is useless. I see almost 0% chance of people
> > taking the docs and doing something with them we'd object to.
> If there are no restrictions, someone could copyright a modified version of
> the manual, making it proprietary, and then sue Linux distributors for
> copyright infringement. I agree that the chances that to happen is almost
> 0%, but who knows...
Sounds overly paranoid to me.
> > OTOH I see a lot of impediment to people advertising their work properly if
> > we have too many restrictions on the docs. If someone has built a
> > substantial interface built on plplot/Tk, I think they should be free to
> > republish the docs as their situation requires.
> What situation could that be, realistically? If we are talking about
> company internal reports, I think that the company can do nearly anything it
> wants and we will never know about it.
In general a company is not going to do something that could get them sued,
regardless of our likelihood of "finding out". So if they want to include
the plplot documentation along with their regular user docs, forget it, it's
not gonna happen under a restrictive documentation license.
> However, our licensing terms are
> important if the company is going to make an external publication of the
> modified manual. In this case, I do not see why a company would refuse to
> comply with the proposed licensing terms.
My target example is more like a Japanese govt agency, for which the
bureaucratic rules become more complicated. Just the addition of a simple
cover sheet complicates matters. It's not worth it..
I just don't see the gain from an overly restrictive documentation policy.
The docs exist to serve the program, and have no purpose otherwise. It makes
perfect sense for them to be distributable under a much weaker policy.
Lightspeed Semiconductor Corp