Request for Comment

Rick Moen rick at
Tue Jan 23 04:09:22 UTC 2007

Quoting Ben Tilly (btilly at

> Rick Moen is the most prominent representative of those who believe
> that OSD #6 should be interpreted as forbidding restrictions whose
> effect is to render code unpalatable for certain purposes....

This fundamentally misrepresents what I said.  Thus, I find myself in
the undesired position of clarifying yet again.

As a reminder, in that subthread, at that time we were discussing
MuleSource's MPL 1.1 + "Exhibit B" licence -- not GAP.  I pointed out
that the licensor's _explicitly_ avowed intent (in its Exhibit B clause)
is to make usage of the covered code so impractical for third-party
commercial users that they would be driven to purchase separate,
different commercial-use licences for any such usage models.

I pointed out that OSD#6 is, as clarified by among other things the
posted rationale in the OSD Web text, specifically aimed against
licences that impair usage in commerce.

Plainly, not all conceivable mandatory-advertising clauses would be
fairly interpreted as likely to impair use in commerce.  (I suggested
one such possible clause.)  Plainly, MuleSource seeks explcitly to do
so:  The sponsoring firm has openly admitted that precise intent.

Concerning GAP in distinction to MuleSource's "Exhibit B", I pointed out
that a licensor invoking its wording...

   a display of the same size as found in the [original code] released by
   the original licensor

...could require all derivative works to sport a 500-point logo +
company name + URL display, specifically to make commercial use
impractical.  I.e., the lack of any limit on size and promience
(completely aside from the OSD#10 issue) provides a method for licensor
to effectively prevent competing commercial use.

You believe that doesn't contravene OSD#6 because commercial competitors
merely don't WANT a 500-point competitor's logo, not that they CAN'T --
an argument I consider self-parodying and make no further comment.

> However one gets the same effect from taking any pair of incompatible
> licenses and joining them together.

This, however, is not a joining of two codebases under incompatible
licences at all, but rather a defect in a derivative work formed
entirely from code all under _one_ licence -- so your comment is

I'm disappointed but not surprised that I'm having to repeat myself on
this matter yet again.  Please do not make necessary yet more iterations.
Thank you.

More information about the License-discuss mailing list