Copyright issue (was: Re: bug#19479: Package manager vulnerable)

[Kelly Dean <kelly@prtime.org>]

Richard Stallman wrote:
>> The following say that's an option:
>> http://git.savannah.gnu.org/cgit/gnulib.git/plain/doc/Copyright/disclaim.manual
>> http://git.savannah.gnu.org/cgit/gnulib.git/plain/doc/Copyright/disclaim.changes.manual
>> http://git.savannah.gnu.org/cgit/gnulib.git/plain/doc/Copyright/disclaim.program
>
>> The copyright clerk declined my offer.
>
> Would you please show me the mail you exchanged?
> If we handled this wrong, I'd like to see what we did wrong.

Sure.
On Dec 19, 2012, Donald R Robertson III via RT <copyright-clerk@fsf.org> wrote:
Re: [gnu.org #787588] Copyright paperwork for Emacs contributions
[snip]
> I have included 
> an assignment form for you. If you are employed to do
> programming, then please also have the disclaimer 
> included at the bottom of this email signed by your
> employer.

On Dec 23, 2012, I replied:
[snip]
> I'm an American citizen and resident, and have no employer or school that can claim my work. I disclaim copyright on my work, so there's no copyright to register. It's in the public domain.

On Feb 19, 2013, he replied:
> In the four and a half years I've
> been working here, I have only had one 
> disclaimer from a contributor. It really is more beneficial
> for our enforcement efforts if we get the 
> work assigned instead of 'disclaimed'. We will only accept a
> disclaimer instead of an assignment in 
> particular circumstances.

On Feb 20, 2013, I replied:
> I disclaim copyright ownership, so there's nothing to assign. Fortunately the few patches I've submitted so far are under the 15-line limit that you accept without copyright assignment, except for one unimportant feature enhancement that hasn't been applied yet, so the lack of copyright assignment won't cause problems for the maintainers. I'll try to structure my future work to be add-ons rather than part of Emacs, so the copyright issue won't matter.

That was the last I heard from him. I'm CCing this message to assign@gnu.org, and the clerk can send you full copies of all the messages we exchanged, including stuff about his broken spam filter (which is why it took him almost two months to respond), etc, if it matters.

Because the FSF had no disclaimer or assignment on file for me, Glenn Morris asked me not to submit any more patches (not even trivial, one-line ones) to apply to Emacs. I abided (and abide) by his request, including for bug #19479. Stefan then invited me (mistakenly, it turns out) to submit a patch for that bug, so I did. I also submitted a trivial one-line patch to fix bug #19536 because of its effect on validation of the former patch.

Glenn then objected. Since I'd already spent time to study package.el, develop my patch, test my patch to make sure it solved the problem that I said it solved, test it to make sure that the file format extension it introduced wouldn't interfere with old (non-patched) clients, and test it to make sure it worked with both Emacs 24.4 and the latest trunk, I reminded Glenn that my work is PD, that Emacs already includes PD code (feedmail.el), and that all of my contributions to Emacs amount to an extremely small part of the codebase, so including my PD code would have no more effect on your ability to enforce the GPL than would including any other PD information, such as a few lines of Hamlet, and I recommended that he not reject my patch just because it's PD. And I asked him not to conflate 
 the issue of the FSF refusing new PD code with the fact that my code is PD.

Stefan then said the problem was that I hadn't signed the assignment form. That's why I pointed out that the FSF says it accepts either assignment or disclaimer, and bothered to fill out, sign, and submit a disclaimer yesterday even though the clerk refused to ever send me one to sign. I hope this puts to rest the question of whether my code is PD.

Stefan still insists that my code is not PD, and that it won't be until some years after I die. His claim contradicts Emacs's own source code (feedmail.el) and GNU's own website, as well as SQLite, the PD status of which is indisputable.

Obviously, I can't assign ownership of PD code to the FSF. The FSF is welcome to use PD code, including both feedmail.el and the patches I've submitted. I have no control over whether the FSF chooses to do so.