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

Re: summary of software licenses in non-free



Scripsit Wouter Verhelst <wouter@grep.be>
> Op di 13-01-2004, om 00:48 schreef Anthony DeRobertis:

> > >> good idea.  perhaps something easily parsable like:
> > >> Non-DFSG: 1, 3, 5

> > I don't think it'd be sufficient to do that with. DFSG 3, for example, 
> > is _very_ broad. 

> Yes, I know. So is 5, which makes it a lot less efficient for the
> purpose I suggested. Still, that was only an example; and if it is to be
> implemented, it should have an advisory character, at best.

The problem is that it would be hard to make use of such a line
without confusing uninitiated users. For example, if a package in
non-free had

 Non-DFSG: 3

and a tool that parsed that displayed

 This package is non-free because
   - it does not allow modifications and distribution of modified source.

someone might complain that the license does allow you to distribute
modified sources, if only your stand on your head and chant the name
of the author's cat thrice. It's hard to describe what is wrong with
this simply by pointing to entire clauses in the DFSG.

Or imagine a license that was all fine and dandy except it contained a
"you must monitor my website" clause. TTBOMK all debian-legal regulars
agree that we do not consider that free, but which clause of the DFSG
would we point to here? The best we could do would be to put

 Non-DFSG: 1, 2, 3, 7

(i.e. all the clauses that positively require rights), because rights
that are granted only to people who monitor websites do not count when
applying the DFSG. However, such a line would not make the reader much
wiser about what was *really* wrong with the license.

> But having a machine-readable list of (generic) reasons why a certain
> piece of software is non-free has its advantages. We should do that, if
> possible (and it is possible).

It might well be the case that we could construct a workable
classification of "errors" (wrt the DFSG), but I'm disputing the
assumption that the clauses of the DFSG itself consitute a good
attempt a such a classification.

A better start might be the list of common fallacies in Q11 of the DFSG
faq on <http://people.debian.org/~bap/dfsg-faq.html>, but it would
take some additional work to structure it as a good classification
scheme.

-- 
Henning Makholm                          "Hell, every other article you read
                                      is about the Mars underground, and how
                           they're communists or nudists or Rosicrucians --"



Reply to: