[License-review] For Approval: OpenMDW License Agreement, versions 1.1 (OpenMDW-1.1)
Rob Landley
rob at landley.net
Sat Aug 15 17:17:24 UTC 2026
On 8/13/26 14:00, Pamela Chestek wrote:
> I recall a presentation at a conference many years ago tracing the
> history of different licenses and the scope of revocation. As I recall
> it, there was an expansion to include copyright termination but then
> that retracted back to just patent. Can anyone provide some color on the
> thinking about why copyright termination may or should be disfavored
> upon a patent infringement claim?
In a proper open source license, there's no "one true owner". Anybody
can fork the project and maintain their own version under the same
license. I believe failure to allow that is against the open source
definition.
Somebody forks a project and bolts on proprietary code. If the
proprietary company sues the fork, do they lose access to the _upstream_
version of the project?
That's aside from upstream being a bad actor. (If the linux kernel was
under this license, Linus could take anybody's code and they could do
nothing about it: an obvious attempt to create Too Big To Sue.)
Also, if somebody takes _this_ project's code and puts it under a
different license... can the project _itself_ sue? Can the license EVER
BE ENFORCED? Because it would terminate its _own_ right to use the
licensed material. (Unless it plans to never have external contributors
and thus not need its own license. And even with no external
contributors, could anyone else safely use a license issued by someone
that had explicitly terminated its _own_ ability to use that license? I
can see side-eye from a judge about splitting that hair, let alone
claiming the resulting "no, I'm magic and special, only I can do this"
has anything to do with "open source".)
But if you want to approve a license that can either never be enforced
or only enforced by a specific custodian in whom all rights and
ownership vests nontransferrably... you do you.
Rob
P.S. this is just the surface level stupid. There's almost certainly
more stupid if you dig.
More information about the License-review
mailing list