Re: Response from MySql AB (Re: Humor me: Postgresql vs.

"Marc G. Fournier" <scrappy(at)postgresql(dot)org> writes:

> On Thu, 9 Oct 2003 jearl(at)bullysports(dot)com wrote:
>
>> Andrew Sullivan <andrew(at)libertyrms(dot)info> writes:
>>
>> > On Thu, Oct 09, 2003 at 08:52:36AM -0600, scott.marlowe wrote:
>> >> Fact: If you write your application to work with ODBC -> MySQL
>> >> connectivity, you can write a closed source app and sell it for money and
>> >
>> > Fact: nobody's ever tested any of this in court, so you're basically
>> > risking it.
>> >
>> > I think if people want legal advice about the status of MySQL's
>> > claims about GPL, they'd best consult a lawyer who knows a lot about
>> > software licenses.
>>
>> This is especially true considering the fact that the Free Software
>> Foundation would *love* for MySQL AB to be right on this one. The
>> interpretation of the GPL the FSF forwards is the one that they
>> feel that they can safely defend in a court of law. However, if
>> there was a precedent set for MySQL AB's interpretation that would
>> suit them right down to the ground.
>
> 'K, you lost me here ... from what I've seen, MySQL AB's license is
> "GPL with exceptions that force you to use our commercial version"
> ... whereas my understanding of the GPL itself is that there are no
> exceptions, period ...

I can see why I lost you. I left out the most important bit. MySQL
AB uses the same GPL as the rest of the world. The difference is that
MySQL AB's interpretation of the GPL relies on a very broad
interpretation of "derivative work." They tell their customers that
if the application requires MySQL that it is a derivative work
regardless of all the esoterica like how the application might be
linked, etc. From what I understand there are actually some
precedents that could be stretched to support this position.

The definition of a derivative work that the FSF uses is considerably
more narrow, but that's not because they want it to be so. The FSF is
simply using a definition that it is relatively certain that it can
defend in court. If MySQL AB were to set a precedent that widens what
constitutes a derivative work then the FSF would be very pleased.

Most developers don't have a clue about the intricacies of Free
Software licenses, and most managers are wary about getting on the
wrong side of the GPL (I would be too if there was a chance of having
to face Eben Moglen in court). Since the commercial license for MySQL
is not too terribly high, MySQL gets quite a few sales. Since the
whole kit-and-kaboodle is untested in court you can't even really
accuse MySQL of being devious. It's even possible that the courts
will uphold their version of what constitutes a derivative work.

Those of us here would rather go with the altogether safer PostgreSQL.
It won't truncate your large integers without warning, it doesn't
consider 0000-00-00 a valid date, and it's license is unambiguous.

I hope that is a little clearer.

Jason