Re: GPL v2 authoritative answer on stored code as a derived work

Jonathan Smith [off-list ref] writes:

It's pretty clear that code stored in a Git repository isn't
considered a derived work of Git, regardless of whether it is used
in a commercial context or otherwise.

However, I'm unable to find this stated in any authoritative and
unambiguous manner.

Is it reasonable to ask for such a statement?  I doubt it,
especially if "It's pretty clear".

Without such a statement, I think we have already seen that the
commercial adoption is already appealing.

From: "Junio C Hamano" <redacted>

Jonathan Smith [off-list ref] writes:

quoted

It's pretty clear that code stored in a Git repository isn't
considered a derived work of Git, regardless of whether it is used
in a commercial context or otherwise.

I'm guessing here, but I suspect that while its 'pretty clear' to Jonathan, 
that he has met others who aren't so clear or trusting, and it's that 
distrustful community that would need convincing.

quoted

However, I'm unable to find this stated in any authoritative and
unambiguous manner.

The GPL2 FAQ's (search for 'data')
http://www.gnu.org/licenses/old-licenses/gpl-2.0-faq.en.html#TOCGPLOutput 
(should link to "Is there some way that I can GPL the output people get from 
use of my program? For example, if my program is used to develop hardware 
designs, can I require that these designs must be free?") and 
http://www.gnu.org/licenses/old-licenses/gpl-2.0-faq.en.html#IfInterpreterIsGPL 
both cover the fact that the repo (data) is not a derived work, and thus not 
subject to GPL2.

Is it reasonable to ask for such a statement?  I doubt it,
especially if "It's pretty clear".

(Rhet) If there were to be such a statement, where should it be placed, and 
who could issue it? It couldn't be part of the COPYING licence file (because 
it's not supposed to be modified).

It could be in the User Manual, but that wouldn't carry much weight with the 
already worried, or perhaps the git(1) man page [E.g. 'discusssion' section 
maybe], or even in the git-scm.com 'book', but really it would need Jonathan 
(and others with similar FUD issues) to suggest what they'd need.

Without such a statement, I think we have already seen that the
commercial adoption is already appealing.

Hopefully the links can help Jonathan if he is having any local 
difficulties.

--
Philip 

Hi Philip,

On Wed, 27 Jan 2016, Philip Oakley wrote:

From: "Junio C Hamano" <redacted>

quoted

Jonathan Smith [off-list ref] writes:

quoted

It's pretty clear that code stored in a Git repository isn't
considered a derived work of Git, regardless of whether it is used
in a commercial context or otherwise.

I'm guessing here, but I suspect that while its 'pretty clear' to Jonathan,
that he has met others who aren't so clear or trusting, and it's that
distrustful community that would need convincing.

It is not so much distrust as something you could take to court, I guess,
because an *authoritative* answer was asked for. Now, the question is a
legal one, so it is pretty clear (;-)) to me that only a lawyer could give
that answer.

Having said that, I know of plenty of companies storing their proprietary
code in Git repositories, and I would guess that they cleared that with
their lawyers first.

Jonathan, please do not take that as any indication that I try to give
this answer: if you want an authoritative answer to your question, you
really need to consider asking a lawyer.

Ciao,
Dscho

From: "Johannes Schindelin" <redacted>
Sent: Thursday, January 28, 2016 8:16 AM

Hi Philip,

On Wed, 27 Jan 2016, Philip Oakley wrote:

quoted

From: "Junio C Hamano" <redacted>

quoted

Jonathan Smith [off-list ref] writes:

quoted

It's pretty clear that code stored in a Git repository isn't
considered a derived work of Git, regardless of whether it is used
in a commercial context or otherwise.

I'm guessing here, but I suspect that while its 'pretty clear' to 
Jonathan,
that he has met others who aren't so clear or trusting, and it's that
distrustful community that would need convincing.

It is not so much distrust as something you could take to court, I guess,
because an *authoritative* answer was asked for. Now, the question is a
legal one, so it is pretty clear (;-)) to me that only a lawyer could give
that answer.

Having said that, I know of plenty of companies storing their proprietary
code in Git repositories, and I would guess that they cleared that with
their lawyers first.

I've had a look though the various FAQs and other discussions about the
GPL and the FUD associated with it.

I've put together an outline of a patch to the git(1) man page, with commit
message to explain the issues (the lawyers need pointing in the right 
direction
so they can think clearly, rather than give the usual 'No' answer ;-)

Having it in $gmane at least captures the rationale, even if the patch goes
nowhere.

Jonathan, please do not take that as any indication that I try to give
this answer: if you want an authoritative answer to your question, you
really need to consider asking a lawyer.

Ciao,
Dscho
--
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Some potential commercial users may be put off by the FUD
(Fear, Uncertainty and Doubt) that has been raised in the past
regarding the use of FOSS software.

In such communities, even the legal advice may be misinformed
and over-cautious regarding the storage of code and other
intellectual property in a Git repository for fear that Git's
GPL2 licence may somehow 'infect' the respository.

Add simple statements highlighting Git's licence, it's use
for Linux, to imply industrial-strength, and that users should
apply a suitable licence of their choice because the Git GPL
licence does not apply to their repo data.

It should be noted that a 'git init' will create a repo that while
empty of user data does provide the .git directory structure, which
includes a number of template files ('hooks\pre-rebase.sample' is
explicitly copyright), default refs and config file. Some may suggest
that these carry the GPL2 to the repo.

The GPL2 will still apply to the hook templates and the other
template files, but these, even if modified (becoming derived works)
would be distributed with the repo, satisfying the GPL.

The new content copyright belongs to user. Request that they state
their licence terms in line with recent FOSS industry practice.

Signed-off-by: Philip Oakley <redacted>
---
asciidoc formatting not checked.

The fear of 'infection' of a repo by the templates copied to the
repo by 'git init', should not be underestimated, given the need
for the Bison exception. Sometimes it has to be spelt out why it's
not an issue (in the commit messsage)

https://help.github.com/articles/open-source-licensing/

http://thread.gmane.org/gmane.comp.version-control.git/284715/
---
 Documentation/git.txt | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/Documentation/git.txt b/Documentation/git.txt
index bff6302..137c89c 100644
--- a/Documentation/git.txt
+++ b/Documentation/git.txt

@@ -1132,6 +1132,17 @@ of clones and fetches.
 	  - any external helpers are named by their protocol (e.g., use
 	    `hg` to allow the `git-remote-hg` helper)
 
+Licencing: Your data, and the Git tool[[Licencing]]
+---------------------------------------------------
+
+Git is an open source tool provided under GPL2.
+Git was designed to be, and is, the version control system
+for the Linux codebase.
+Your respository data created by Git is not subject to Git's GNU2
+licence, see GPL FAQs
+http://www.gnu.org/licenses/old-licenses/gpl-2.0-faq.en.html#TOCGPLOutput).
+
+User should apply a licence of their own choice to their repository data.
 
 Discussion[[Discussion]]
 ------------------------

-- 
2.7.0.windows.1