On Fri, Feb 10, 2006 at 12:33:05PM +0000, Roger Leigh wrote: > Anton Zinoviev <[EMAIL PROTECTED]> writes: > > > > I think the following is an useful test. If the license forbids some > > modification that is necessary in order to adapt the document to some > > need, then the document is non-free. Otherwise, that is if the > > license does not forbid any necessary modification, the document may > > be free. > > This is no good. Where is it defined what is "necessary", and who > deems what is "necessary"? What /I/ consider to be necessary may be > considered "unnecessary" (and hence, not allowed) by the copyright > holders.
I don't think we disagree what "necessary" means. > As an example, the FSF do not appear to consider the ability to remove > invariant sections necessary in the current version of the GFDL for > example, whereas I (and others) do. The reference cards were just an > example of this need; aggregate works were another, The reference cards do not require the removal of the invariant sections. You can print the invariant sections on separate sheet(s) of paper. > and there were several other real-world cases where a need was > demonstrated. I tried to list them in the following link and I don't think that a need was demonstrated in any of the examples. http://lists.debian.org/debian-vote/2006/02/msg00226.html > Applying your test, in my eyes, still leaves the GFDL a non-free > licence. I understand that this seams so, but no example was given to prove this. > Could we draw this debate to some sort of conclusion? I continue to > remain unconvinced by the majority of your arguments, many of which > are still poorly explained. If necessary I can try to explain better. Anton Zinoviev -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
