[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Copyright/licensing action plan + a sample [PATCH]

From: Anthony W. Youngman
Subject: Re: Copyright/licensing action plan + a sample [PATCH]
Date: Mon, 14 Sep 2009 00:29:29 +0100
User-agent: Turnpike/6.07-M (<U1e6TVn8PTCuG3mvbGX+2+ajAT>)

In message <address@hidden>, Reinhold Kainhofer <address@hidden> writes
Am Montag, 14. September 2009 00:00:28 schrieb Anthony W. Youngman:
DON'T track "whether they support switching the licence". Because if
they do, they will (presumably already) have switched the licence on
their contributions.

I don't think so. Many contributors simply don't mind a bit about the license
and the formalities. Coding and the results are more important than dealing
with stupid license headers, etc.

Unfortunately, taking an ostrich attitude with the law is rarely productive...

You are exhibiting a touching, blind, blinkered faith in the FSF. If I
may speak for Han-Wen, I don't think he shares that faith. There may
well be lilypond contributors who don't believe in the GPL, surprising
as that may sound! But there's nothing stopping BSD believers (who may
find the GPL offensive!) from contributing to lilypond.

My understanding has always been that contributions should be GPL v2...

The overall licence is v2, therefore contributions need to be at least v2-compatible.

DO NOT try to switch the licence to v2+. You will probably run into a
brick wall! And if the eventual plan is to be v3-compatible you're
setting yourself up for failure!

I think the eventual plan should be to be GPL-compatible in the long run, so
v2+ would really be best. Otherwise we'll have much bigger headaches one GPL
v4 comes out and libraries start switching to it.

I agree v2+ is best. I just also think people like Han-Wen and Linus have a point. That means that, for some people, the "+" is going to be a non-starter.

Use your spreadsheet to *track* *all* the licences to lilypond,

Please, do we really need a law firm to keep up lilypond development?

Nope. But being an ostrich doesn't help... if we don't know what licence individual contributors have used, we're going to have problems when v4 comes out...

restrict the licences you can handle to an arbitrary subset of the
licences you think other people should use (that attitude is offensive).

Huh? We are a GNU project, and the guidelines of GNU are GPL.... (Which
doesn't mean that everyone is forced to use GPL, but that the standard license
SHOULD be the GPL). so I don't see the v2+ discussion as offensive, but as the
only sane choice that ensures that in 5 years down the road lilypond will
still be able to use up-to-date tools.

v2+ is sane FOR LILYPOND for exactly the reasons you suggest. But it is INSANE for some developers because it hands control over MY code to YOU (for various values of "you" - in this case the FSF).

for example, what would we do if freetype changed its license to GPL v4 (their
freetype license is GPL-incompatible, so we rely on their GPL, which is v2,
btw)? We'd be doomed, because we couldn't link to freetype any more. And
tracking down early contributors becames harder each day.

Well, what would we do now? At least if we know that certain chunks of code are v2 only, we can put contingency plans in place.

As it is, I find your emphasis on v2+ offensive, and I doubt I'm alone.
Given the choice of "v2 or v2+", I'd go for "v2 only". But if you ask me
"what licence would *I* choose?", my reply would be "v2/v3". See what I
mean about your approach being counter-productive?

I repeat. Sod *your* choice of favourite licences. Just *track* the
licences contributors have chosen, and then you can also track whether
the licences are v3-compatible.

So your idea is basically postponing the problem to a time, where we might not
be able to solve it properly any more?

Nope. My idea is basically saying "let's face reality. Some people will never agree to "or later" so let's have a contingency in place". Seeing as that seems to include certain MAJOR contributors like Han-Wen, then having that contingency seems to be a very good idea.

Anthony W. Youngman - address@hidden

reply via email to

[Prev in Thread] Current Thread [Next in Thread]