FSF is creating a problem that never existed!
Posted Oct 2, 2006 19:26 UTC (Mon) by
landley (guest, #6789)
In reply to:
FSF is creating a problem that never existed! by mingo
Parent article:
Busy busy busybox
> I'm also worried that RMS justifies the closed decision-making process
> of the GPLv3 because: "Most of our community does not appreciate
> freedom" .
While I remain as amused as ever by those who would impose democracy at
gunpoint, spread by the sword the philosophy of turning the other cheek,
or otherwise force people to be free against their will, I'd like to point
out that the argument being covered here wasn't really about the contents
of GPLv3. It's about GPLv2 being "good enough", and a dual license being
extra work to maintain.
Some people consider a BSD license to be "good enough". I disagree
because a BSD/MIT style license seems to encourage forking by having your
developers hired away to work on a proprietary version (ala the history of
BSD: Bill Joy leaving Berkeley for Sun in 1982, the CSRG going to BSDI a
decade later, Jordan Hubbard going from FreeBSD to MacOS X a few years
ago...) GPLv2 prevents that; the forks can converge because you always
get the patches back, and Linus or Andrew can change employers freely
without jeopardizing their ability to contribute to the project. This is
pragmatically useful.
With GPLv3 the FSF is off on an ideological kick trying to open up the
Xbox, which has _NOTHING_ to do with the purpose of GPLv2. I see the
purpose of GPLv2 as to keep open source projects from forking off closed
source proprietary versions that suck away seasoned developers every time
they develop enough momentum to become commercially interesting. GPLv2
keeps code commodity, and prevents anybody from trying to corner the
market by throwing money at developers. (Maybe this is bad for some
individual developers, but it's very good for the _project_ and thus
better for _most_ developers. And Linus hasn't exactly gone hungry.)
GPLv2 has an excellent pragmatic effect, good for the project's
development. And when shooting for this effect, GPLv2 nails it. Thus I
continue to want to use GPLv2, no matter what the FSF says. Now the FSF
seem to be pissed that Google won't let users run arbitrary code on its
servers, and are trying to draft me into this new fight over Tivo and the
Xbox (I own neither), and I'm just not _interested_. That's not what I
thought this "or later" clause was for, I dislike being coerced, and I'm
not giving them any more blank checks.
But fundamentally, the central issue for me is that GPLv2 is good enough,
it still works fine for me, I don't need other licenses, and I never asked
for more than GPLv2 provides. There didn't used to _be_ a GPLv3, and I
still don't see the need for it. Using "but what if it goes away"
fearmongering to leverage a whole new agenda is kind of slimy, and despite
the chicken little act GPLv2 is NOT going away as long as the Linux kernel
uses it. And if I have to make a choice between throwing in my lot with
the Linux kernel developers, or throwing in my lot with the FSF? It's no
contest.
(
Log in to post comments)