[Fwd: Re: Commercial Open Source- state of play]

[Tony Butterfield]

On Mon, 2003-09-15 at 08:01, Ian Lance Taylor wrote:
> Sleepycat's definition of ``redistribute'' does not in any way affect
> the open-source status of their license.  What Sleepcat is saying is
> that if you ``redistribute'' according to their criteria, you must
> include the complete source code for your application.  That is a
> standard open-source requirement, found in the GPL among others.
> Sleepycat definition of ``redistribute'' does not remove that right;
> it actually give you more rights (i.e., not distribute source code),
> and thus can not decrease the open-source status of the license.

Thanks for your response and clarifications. Your description made a cog
turn inside my head. You can get the effect of requiring closed source
folks to buy commercial licences (if one is available) by using
non-discriminatory but carefully targeted clauses. I've been pointed in
the direction of the Reciprocal Public Licence that seems even stronger
in this respect, the Apple Public Licence and the Open Software License
to a lesser degree.

> 
> > 2) Dual licencing approaches allow revenue to be created from
> > distributees who want to avoid the terms of an open licence. But this
> > seems only useful for distributees who want more flexibility as
> > distributors. It couldn't really work for "end-users" as they can pretty
> > much do as they require under the terms of an OSI licence as long as
> > they don't distribute. Is this a fair understanding or are there more
> > subtleties that I have missed?
> 
> I believe that is a fair understanding.  Most people feel that open
> source licenses may not restrict running the program.

The Reciprocal Public Licence also requires that you open up your source
if you deploy the functionality of a derived work that is available to
anybody else. Clauses like these require even "end-users" to be open
source or buy a commercial licence.


--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3