When a product is a derived product?
While working on my little wiki on yardradius, I found something which
could be of interest for this ML. Cistron Radiusd 1.6 is released under
GPL, and pkg description says:
This GPLed Radius server is not based on any Livingston code. It is
compatible with the Livingston-2.01 server though. Over radius-2.01,
it has support for Exec-Program on authentication, it is possible to limit
the number of concurrent logins reliably, it has tagged attribute
support, it can replicate accounting packets, and more.
If you worked like me on radiusd stuff since a few years, you know
that the Cistron product was a clear derivation of the original Livingston 1.16
in the old days. The old product is yet available here:
ftp://ftp.portmasters.com/pub/le/radius/
You can find that some files have yet the same names and some functions too
(along with their arguments and returns types).
Of course, the original product has been largerly rewritten and improved,
a lot of stuff and code added and so on. Anyway, part of the ancient
code is there. The old radiusd was released under a 1992 edition of
a BSD-like license:
* Copyright 1992 Livingston Enterprises, Inc.
*
* Permission to use, copy, modify, and distribute this software for any
* purpose and without fee is hereby granted, provided that this
* copyright and permission notice appear on all copies and supporting
* documentation, the name of Livingston Enterprises, Inc. not be used
* in advertising or publicity pertaining to distribution of the
* program without specific prior permission, and notice be given
* in supporting documentation that copying and distribution is by
* permission of Livingston Enterprises, Inc.
*
* Livingston Enterprises, Inc. makes no representations about
* the suitability of this software for any purpose. It is
* provided "as is" without express or implied warranty.
Now, the obvious question: is the re-licensing under GPL legitimate? My opinion
is not, due to GPL-incompatibily of the old-BSD license.
When a 'derived' product can be considered not (or no more) 'derived'
(i.e. how much of its code should change) by another one?
--
Francesco P. Lovergine
Reply to: