[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [gNewSense-users] Linus' statement about copyright
From: |
Paul O'Malley - gnu's not unix - |
Subject: |
Re: [gNewSense-users] Linus' statement about copyright |
Date: |
Thu, 08 Jan 2009 23:15:26 +0000 |
User-agent: |
Thunderbird 2.0.0.19 (X11/20090105) |
Matthew Flaschen wrote:
Matthew J. Fisher wrote:
This implies that any kernel files, in which Linus and other authors
assert copyright without specifying a license, really are covered under
the GPLv2 as stated in the COPYING file. Anything else would be
illogical.
Also, at Sam mentioned a few weeks ago, copyright always applies --
whether or not it is asserted. So this is not much of a special case.
I agree.
Still, people ARE sometimes illogical. It wouldn't hurt to document
these buglets, and let RMS see if he can obtain confirmation that the
files are free.
Hi,
The correct place to address licensing questions is to:
address@hidden
In the main it is when you find something that is not quite the GPL or
something that is not quite the source.
As for files that are not documented I think like this:
What would a
-- Project Copyright licence --
cover, and it seems to me that it would cover all below.
root of source tree
|
|----(branch)----File1
|
|
FileN ----------------------FileN+1
|
|
Where a project has an explicit copyright, look only for deiviations
from that, and look for items that contradict that (blobs), all else is
a waste of time unless a court of law tells you otherwise (IANAL etc).
It is most likely fair to do this, other things are not hitting targets
we want to hit, when we define what we can't have, then we are left with
what we can have.
While this does not deal with every edge case it is a good general rule.
In reality if you have a good understanding of what you are at then you
would only have to send edge cases to address@hidden
The reason I am suggesting you use something like that rule is that
people behind address@hidden are fairly busy, adding things to their
list that are solvable in other ways.
The easy way to evaluate the software licence is to do this.
Does it pass the four freedoms.
Each one provides a separate test.
If so, then there should be no problem.
If it fails then the question is "how is it non free", with this new
knowledge you could ask the author (very politely) to consider an
alternative or dual licence.
If the code contains other peoples contributions then they too have
rights in there, so they too have to be asked. The best example of this
is the GLX case.
Regards,
Paul