Subject: Re: SocialText license discussion--call for closure of arguments
From: Rick Moen <rick@linuxmafia.com>
Date: Sat, 20 Jan 2007 07:18:51 -0800

Quoting Andrew C. Oliver (acoliver@buni.org):

> I'm discounting all emotional content of messages in this thread and
> focusing on the constructive content.

For Web page summaries and other analysis, that is certainly a good idea.  
I maintain that there is also room, during discussion, for justified
annoyance at transparently and insultingly bogus arguments, especially
from interested parties such as Mr. McIntyre.

By the way, my thanks to you and Matthew for what looks like a good job
of attempting to summarise.  (I'll find time to look more closely, soon.)

> >  Just append an advertising clause to MPL 1.1 so onerous that no 
> >   competitors are going to be able within reason to assert their OSD#6
> >   rights to use the code for any purpose.  And then claim in public
> >   that it's open source, which you'll get away with if you
> 
> Do you think this is a fair argument (OSD #6)?  It seems to me for
> this to be justified the license would have to specify this clearly:
> "Must not be used by FirmX" or "Must not be used to XYZ"

Yes, I think it's absolutely a fair argument.  OSD #6 originated, I
believe, as a reaction in part to the many "free for non-commercial use" 
licences popular in academia prior to the BSD and X11 licences'
emergence at UCB and MIT -- e.g., those of Mosaic from NCSA at U. of
Illinois; COPS (Computer Oracle and Password System), SAINT, and (early)
Tripwire from Perdue University, etc., that put out academic-licensed
"free" products to popularise code, while reserving some or all
commercial rights.

The purport of OSD #6 is that, by contrast, must be freely usable for
any purpose, specifically including commerce on an equal footing.  Thus 
the wording of the "Rationale" on the OSD page:

   Rationale: The major intention of this clause is to prohibit license
   traps that prevent open source from being used commercially. We want
   commercial users to join our community, not feel excluded from it.

("Any purpose", above, is not construed to bar licences that withhold
the right to create proprietary forks, i.e., copyleft licences.)

Now, there is no specific part of OSD #6 or anywhere else in that
document that says "must be usable for any purpose other than
(optionally) creation of proprietary derivatives" -- not exactly.  It's
understood, that being one of the core _ideas_ that OSD is trying to
formalise.  OSI hasn't _yet_ felt it necessary to say in the OSD "must
not be commerce-impaired, as long as it's equally commerce-impaired for
everyone but the copyright holder" -- just as OSI didn't find it
necessary to say "License Must Be Technology-Neutral", that being also
an implied consequence of the OSD's core notions, until things like that
Attribution Assurance Licence came around and clarified the need for
that clause.


So, I would maintain, the key question is whether this class of licences 
aim to impair commerce.  For the "Exhibit B" licences -- those _actually
used_ by the likes of SugarCRM, Zimbra, Alfresco, Qlusters, Jitterbit,
Scalix, Terracotta, Cognizo Technologies, MuleSource, Communiva, Dimdim,
Agnitas, Openbravo, Emu Software, ValueCard, Open Country, 1BizCom,
KnowledgeTree, and Sapienter Billing Software, I think the answer is
unquestionably "yes", and I'm obliged to Nicholas Goodman for pointing out
that they themselves have made this starkly clear:
http://www.nicholasgoodman.com/bt/blog/2006/12/22/badgeware-ceo-to-community-buy-a-commercial-license/

Googman quoted Mulesource executive Dave Rosenberg to that effect:

    So, if you use Mule in your software product and sell it 
    commercially, then you are required to either make a
    licensing deal with us or keep the "power by Mule" logo
    visible. Just as so many other things in OSS are confusing, 
    it appears that this too has created some consternation -- 
    primarily because people want to embed Mule in their products 
    and couldn't quite make sense of how the attribution would work.

    My answer was simple. You make a deal with us for a commercial
    license and then you do whatever you want.

So, the idea is "Well, sure you could in theory sort of use our code 
in commerce, but we want to make it so unattractive to you that you'll
come back to us and buy a separate commercial-use licence at our asking 
rates -- if we're offering one at all."  That's the Mosaic / COPS / 
Tripwire / SAINT licensing model, slightly softened and updated for the
new millennium.  It's not open source.


That's the "Exhibit B" licences (that are being actually used).  The
separate question remains of the Generic Attribution Provision patch 
paragraph, as applied against MPL 1.1 or one of the other 57
OSI-approved licences.  (Again, Socialtext did not submit a _licence_,
so by implication, in theory we have 58 patch cases to consider.)

Reasonable people might differ as to whether GAP's requirement
substantively impairs commerce for all but the copyright holder.  
(I find it interesting that nobody actually _uses_ GAP + any licence, 
yet:  Intalio proposes to use MPL 1.1 + GAP, but hasn't yet done so.)

   Redistributions of the [original code] in binary form or source code
   form, must ensure that each time the resulting executable program, a
   display of the same size as found in the [original code] released by
   the original licensor (e.g., splash screen or banner text) of the
   original licensor's attribution information, which includes:

(I join others in wondering where Ross Mayfield's verb escapted to, 
from that "display of the same size" clause:  Mutatis mutandis, it's
good for licence phrasing to be in coherent English.)

Notice that the GAP text (above) doesn't actually place any limit on the 
intrusiveness of the "display":  It must be of "the same size".  If
that's a corporate logo in 500 point type, then all derivative works
must include it unchanged.

I submit that that clause as stated -- making allowance for the mangled
English -- _does_ greatly impair commercial use for downstream users,
for that reason alone.

It would be interesting to see a rewrite of GAP -- perhaps one in whole
declarative sentences -- that more arguably doesn't impair commercial
use.  (There's nothing _inherently_ wrong with the idea of a mandatory 
advertising clause, per se.)

-- 
Cheers,                    Higgledy-Piggledy / Kibo Ubiquitous, 
Rick Moen                  Greps for his name in the / Happynet spool.
rick@linuxmafia.com        Interdimensional / Cyberspace deity:
                           Didaktyliaios / Dada is cool. -- Lewis Stiller