On Mon, Jul 19, 2004 at 11:42:26AM +0100, Matthew Garrett wrote: > My rule of thumb would be that any technical restriction that > significantly reduces the set of people who can distribute modified > versions is probably non-free. For instance, patch clauses are fine > since anyone able to modify the code should be able to produce a patch - > a piece of code that required to to reverse engineer a large set of > obfuscated code in order to be able to generate a key from your modified > code to run it would be non-free.
I believe freedom is more than just being able to distribute modified versions; it also means not having to jump through onerous hoops to do so, and being able to fork a work and reuse code without jumping through similar hoops. I might agree with your rule of thumb, but not with its converse--a technical restriction that does not significantly reduce the set of people who can distribute modified versions is not necessarily free. (Even if patch clauses don't run afoul of this particular rule of thumb, I still think they're not free.) > >Making freedom harder to assert is restricting freedom. > > The GPL makes it harder to assert freedom because you need to spend more > time investigating subtle license interactions. Making glib statements > is all well and good, but we accept all sorts of things that make > freedom harder to assert. Indeed, we allow some restrictions on freedom. This was a simply a disagreement with "It being difficult to do that thing does not restrict my freedom, it merely makes it harder to assert it." Restrictions can make it harder to assert freedom, and I think all such restrictions should be scrutinized carefully. -- Glenn Maynard
