[PROPOSAL] Open Source certification

Havoc Pennington rhp at zirx.pair.com
Sat Apr 10 06:14:58 UTC 1999


On Sat, 10 Apr 1999, Will Lowe wrote:
> 
> I agree.  But how?  At the moment there's considerable fragmentation among
> the Free Software community.  And asking for public consultation results
> in utter chaos.
> 
> 	 ... the only way it worked the first time (Mozilla) was to ask a
> few select people (Debian/Bruce, ESR, some others) to coordinate.
> 

Not true. Netscape created a special newsgroup for discussion of the
license which anyone could join. There was discussion; a solid half was
unproductive gnu.misc.discuss spillover, another quarter was God knows
what, but a quarter of it was interesting and/or useful. Netscape people
participated in the discussion, as did people like RMS and Bruce, and
ideas from it were folded in to the NPL/MPL drafts (several of these were
released). Netscape even sent all the newsgroup participants cool Mozilla
T-shirts a few months later, which was a nice gesture.

In the end, almost all of the little nitpicks people could think of were
taken out of the license (one more was recently removed, also in response
to concerns from the public). The original NPL was not free software, the
final draft was. And everyone felt really good about it, because the
process was open. It is not even relevant whether the public conversations
"mattered" or not, though I believe they did.

I have no doubt that there were also a lot of private conversations going
on, within Netscape and between Netscape and ESR and Netscape and Bruce
and etc., etc. Private dealings even started well before public
discussion, IIRC, though my memory is fuzzy on the details. These private
discussions played a big part, no doubt, and there's nothing wrong with
that.

But there was *also* an open, public process.

Since we ("the free software community," which SPI hopefully represents)
own the Open Source mark, we have the right to insist on such a process.
I'm sure the suits value the element of surprise and would prefer
pre-endorsement behind closed doors. But IMO this is an important element
of keeping the community together and functioning healthily. If we lose a
suit or two, so be it. Their loss; we are winning now anyway.

Besides, they can always get pre-endorsed by resisting the temptation to
create an incompatible, incomprehensible, self-branded license that keeps
the code from being reused in other software. As many have pointed out,
Apple's code isn't good for a thing besides running Apple's proprietary
OS; due to the license, it can't even be recycled in any of the free OS's.
This may explain some of the bitterness, apart from any OSD violations.

New licenses should be discouraged, and requiring a public discussion
period is one way to do that (while also ensuring the legitimacy of any
that are created, and the harmony of the community).

Havoc









More information about the Spi-general mailing list