Re: GPL v.3?
So far, we've been able to address these issues more-or-less adequately as
they've come up, but it would be a lot easier if the GPL were protected
by a license of the sort you recommended for standards...
I agree that treating the GPL like the other documents might help
persuade some people. These cases are similar in their structure, and
many hackers tend to assume that cases which are similar in structure
should be treated the same way.
But that principle is not valid. Cases with the same structure can be
very different in the practical consequences of the alternatives, and
then they ought to be treated differently. That is what happens here.
It would be a very bad idea to encourage people to *use* modified
versions of the GPL. Use of different incompatible licenses creates a
serious problem, one which does not occur for incompatibility of
anything other than licenses.
If someone modifies a software standards document into a manual, that
doesn't cause anyone any trouble.
If someone modifies a software standards document into an alternative
incompatible spec, and releases free software to implement that spec,
you may be unhappy he did not support the usual standard. If you care
strongly enough, you can modify the program to implement the usual
standard. So even though he did not do exacty what you want, you are
not screwed.
But if someone modifies the GPL and releases a program under the
modified GPL, that is an incompatibility which you can't correct. You
can't modify his program to put it under the real GPL!
As it happens, people *are* allowed to write their own licenses as
they wish; they can use similar language to the GPL if they wish. At
least under US law, copyright cannot be used to restrict the use of
legal language in things such as licenses. So people can and
occasionally do write incompatible kinds of copyleft using language
much like the GNU GPL. But they surely do it less often than if we
invited them to do it, and that is very important.
Reply to: