Re: xemacs wiki

i feel very uncomfortable with this "well, yes, that's illegal but it's ok in
practice".  kind of thing the police love because it gives them essentially
infinite power.  "in practice" may apply to the specific situation they're
looking into (manuals quoting code), but given that these are general licenses,
they are invariably going to be spread to other situations.  even in this very
situation:  when you're dealing with *window systems*, code can often be much
more.  such a manual might well quote 80 lines of code to show an example of a
basic dialog-box handler routine. etc.

and of course there's the example of someone posting a modified version of
sample.ini on the net and saying, i need some help figuring this out.

i'm more and more starting to believe that most wiki's should be public domain.
wikipedia might have a legitimate concern about large encyclopedias stealing
their work ...  but i'm not convinced that this would actually succeed well, for
the same reasons as all the other non-virus-license products that outlived any
competitor's attempt: the mass of volunteers would continue to update wikipedia,
vs. the huge amount of money required to pay people to update the entries for
hand if you went the proprietary route and really wanted to add enough value
that people would care about your version not just the regular one.

----- Original Message ----- 
From: "Eli Zaretskii" <[email protected]>
To: "Hrvoje Niksic" <[email protected]>
Cc: <[email protected]>; <[email protected]>; <[email protected]>;
<[email protected]>
Sent: Friday, November 14, 2003 10:30 AM
Subject: Re: xemacs wiki


> > From: Hrvoje Niksic <[email protected]>
> > Date: Fri, 14 Nov 2003 16:47:55 +0100
> > >
> > > Since you don't _have_ to use invariant sections in your manuals,
> > > you don't need to be worried about this.
> >
> > But you do have to use them if you're inheriting documentation someone
> > else wrote.  This is not the case with the code, where you can change
> > whatever you like (as long as you follow the requirements of the GPL,
> > such as documenting what you've done and releasing the result under
> > the GPL).
>
> Right, and to some people that is, AFAIU, the most problematic aspect
> of the GFDL.
>
> > The apparent incompatibility between the code license and
> > the documentation license might become a real problem when there is a
> > need to include non-trivial examples in the documentation by copying
> > them from real code.
>
> Yes.  However, as I said elsewhere in this thread, GFDL proponents
> claim that in practice, this isn't a problem, since normally a manual
> doesn't include large fragments of code.  I don't know whether this is
> true, as I personally didn't have any experience either way.  The GNU
> Emacs manual, perhaps miraculously, doesn't have any problems in this
> area.