Re: Copyright/Distribution questions (Emacs/Orgmode)

[Subhan Tindall]

On Mon, Mar 11, 2013 at 7:33 PM, Stephen J. Turnbull <address@hidden> wrote:
> Subhan Tindall writes:
>
>  > Ah, I may see your error here Jambunathan.  Copyright becomes attached
>  > to a work the moment in time it is created (at least in the US),
>
> In any jurisdiction implementing the Berne Convention.
True, my assumptions here are that US law are the dominant ones in this case.
>
>  > and publication has no bearing on it's existence or assignment.
>  > The assignment of rights for "changes and enhancements to the
>  > program " <insert progname here> covers the rights to created
>  > material from *the moment the code is written*,
>
> No, that's false.  Copyright law knows nothing of whether the material
> was written "to be part of", or even contains parts of, Emacs[1], and
> therefore a generic assignment cannot cover code until it is
> contributed *to* Emacs, explicitly by the author inserting a "part of
> Emacs" statement, explicitly by substituting the FSF for himself in
> the copyright notice, or (perhaps, I'm not sure what would happen if
> you maintained your own copyright notice in this case) implicitly by
> committing it personally to a repository of code (not necessarily a
> VCS, but any archive) that is considered "part of Emacs."  (You are
> correct in that distribution of the code or presence in "the official"
> Emacs repo are not necessary, of course.)
>
> For example, the FSF has no claim on my ~/.xemacs/init.el, though it
> contains generic enhancements to XEmacs (the code base for which my
> assignment was explicitly designated) that I will probably contribute
> in the future.
>
> On the contrary, I could write an accounting program in 6502
> assembler, send appropriate documentation to the FSF copyright clerk
> indicating that I consider it to be part of (my version of ;-) XEmacs,
> and my assignment for that program would take effect.
>
> I don't claim that either of these extreme examples is at all similar
> to the cases of ox-html and ox-odt.
Again, all true, *copyright* law does not hold any concept of intent.  However,
the *assignment of copyright* in the FSF contract does cary an
implication of intent. The 'reasonable person' standard
of determining intent would be potentially relevant here as follows:
1) work is created - copyright (in US) immediately comes into existence
2) previously signed FSF copyright assignment form potentially
transfers copyright for works created
to be included in X
3) question: was work created with intent to be included in X? if so,
than transfer in 2) applies
4) if disputed, one important test to apply is the 'reasonable person'
standard - what would a reasonable person
judge the intent in 1) to be?
Example.
I sign a copyright assignment form in order to work on project X. I
then write code to enhance or fix errors in X.
A reasonable person would most likely determine that my intent was add
this code to project X, therefore
my copyright was transferred upon it's creation as part of the
contract I signed.

>
>  > A similar situation is a work made for hire.
>
> Yes, it is similar to a work made for hire in that the scope of the
> work for hire is specified, either in a standalone contract, or by
> order of your employer.
>
> Jambunathan is claiming that he has not yet designated this work as
> within the scope of Emacs, but he may be ignorant of the legal
> implications of committing code to certain repositories.  On the other
> hand, a court might construe his ignorance to mean that no intent to
> contribute was present.  I think that's strained; at the present time
> org-mode code is "tracked" to be included in Emacs and I suppose he
> knew that when he committed.  But AFAIK -- IANAL/TINLA -- a court
> *might* be sympathetic to him.
Again, see 'reasonable person' standard.
>
>  > For example, I work on many programs for my employer.  As part of
>  > my contract, all copyright for that work is ceded to my employer.
>
> Correct in the U.S., I believe, but that is an employment contract,
> and a quite different matter, because it covers *your professional
> activities* and the product *of those activities* (in some cases, 24
> hours a day whether on premises or not).  An assignment of Emacs code,
> extant and to be written, to the FSF is *not* an employment contract.
> It is merely a convenient way to perform an indefinite number of
> assignments with one signature (at least, that's what my lawyer told
> me).
Not so different.  The FSF copyright assignment is a legally binding
contract, and should
be treated as such.  It's not just a convenience.
>
>
> Footnotes:
> [1]  Of course the "parts of Emacs" are presumably copyright FSF, *but
> the changes and enhancements are not* (yet).
>



-- 
Subhan Michael Tindall | Software Developer
| address@hidden
RENTRAK | www.rentrak.com | NASDAQ: RENT