I know I'm going to get a certain amount of flames over this
but what the heck. Maybe we can get some good out of it.
Before you break out the flame throwers please make sure you
read at least the last paragraph.
I think some folks may have missed the point on why IBM or
some other large company might want to "clean up the license".
Here are two areas where I feel a cleanup of the PGL might be
in GNU's best interest.
1) Suppose someone from Deep Pockets Inc. contributes a fix
for a small bug in the kernal. At some other time someone
from NoName-NoMoney Inc. also contributes to the Kernal
code. If the code contributed by the NoName-NoMoney folks
is found to violate the patent of some third company, what
is to keep that third company from sueing for damages
everyone who contributed and Deep Pockets Inc. getting
hammered legally/financially for something they didn't do?
Deep Pockets Inc. may want to be the PERFECT GNU citizen,
but can they afford the exposure of contributing code to
a large shared project like the Linux kernal? They may
want to contribute time, people and systems, but can they
afford the risk?
Is there any way we could modify the license so that our
friends in Deep Pockets Inc. can limit their exposure to
only the code they contribute and not anything done in
the entire kernal? This would make corporate envolvement
in any large GPL licensed project much easier to sell to
corporate legal staffs. Maybe something like having each
contributer accept exclusive legal responsibility for the
code they contribute. I don't know, I'm not a Lawyer.
2) Suppose Deep Pockets Inc. enjoys a real market advantage
because they have the patent or copyright on some general
use item like the worlds greatest compression algorithm.
They would like to allow the use of this algorithm in
only selected GPL'd projects. They don't want their
competitors to be allowed to use this without paying a
license fee, but they want to give permanent approval
for GPL use on a project by project basis.
Right now release for GPL use is kind of all or nothing.
Code for any GLP'd project can be shared by any other
GPL'd project. While this is a great goal, it is not
helping GNU with Corporate Patent/Copyright holders.
Could we modify the GPL to allow this kind of limited
release? Remember our choices here are we can gain the
right to some limited use or have no use at all.
If our friends in Deep Pockets Inc. have no intention of
ever marketing a product for which GPL'd ProjectX wants
to use the algorithm, they could grant that project a
permanent Right to Use their Patent or Copyright but still
not allow GPL'd ProjectZ to use it, because ProjectZ is in
direct competition with Deep Pockets Inc's main cash cow.
As an example of this case, let's say Deep Pockets Inc. is
a disk manufacturer and they have their compression algorithm
in firmware on each of their disk drives. They want to give
permanent permission to use this compression in the GPL'd
Linux ppp networking code. The only thing they really do
not want to allow is using their algorithm to make compressed
file systems on their competitors disk drive products not even
with GPL'd code. Right now, releasing that algorithm for GPL
use does not allow that type of end use restriction.
These are just a couple areas where "Cleaning up the License"
might do us all good. I'm sure there are many other little
tweeks where we (The Corporate world, GNU, and Open Source)
could mutually benefit. If we can somehow find a way to limit
corporate responsibility/exposure and allow a little bit of
granularity in control of code released for GPL use we could
greatly enhanse the idea of Corporate envolvement and active
assistance/participation in GPL'd projects.
My intention here is not to argue for any particular changes
to the GPL. Even the above examples are probably not the best
choices. But what I'd like to do is move this discussion into
more productive tracks. Specifically toword what changes could
be made to the GPL that would serve to enhanse and improve our
ability to attract corporate envolvement, participation and even
funding for GPL'd projects while still not in any way weakening
or diluting the goals of GNU/GPL. I realize this is one very
narrow road with many pitfalls , but can we really say we have
explored ALL the options? The GPL is very good, but is it as
good as it can get?
-- Patrick R. Hickel Senior Computing Systems Administrator Boeing Information and Support Services P.O. Box 24346, Mail Stop 7H-AC Seattle Wa 98124-0346hickel@puffball.ca.boeing.com
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/