Re: New Package for NonGNU-ELPA: clojure-ts-mode

[Lynn Winebarger]

On Wed, Aug 30, 2023 at 10:08 PM Richard Stallman <rms@gnu.org> wrote:
>   > How much of your concern with obtaining copyright assignments is the
>   > developer's willingness to give their own signature, and how much is
>   > around the hassle of needing to involve their employer, even when
>   > their contributions are not at all part of their employment?
>
> First, I have to point out that they are both necessary.  The human
> authors cannot assign the copyright if an employer can legally claim
> it.  By signing the employer disclaimer, the employer waives the
> possibility of claiming the copyright, clearing the way for the
> authors to assign it.

I don't dispute the prudence of the FSF's approach.  I don't believe
the enforceability of employment agreements for at-will employees
claiming assignment of legal rights to the product of work outside of
those produced "in the course of employment" has been established
conclusively.  Unfortunately, the prevalence of arbitration agreements
makes it difficult to see the issue being litigated in a forum that
establishes precedent any time soon.  To me, such agreements should
fall afoul of prohibitions against conditioning employment on
"kickbacks" - which is simply the requirement that the employee give
the employer (or any other person) any "thing of value" (e.g. property
of the employee).

> In practice, cases where the employer won't sign are a tiny fraction.
> Some employers object to parts of the text, but we discuss it with them
> and can usually reach agreement.

I suspect there is a substantial selection bias in the first
statement.  That is, the authors who ask for a copyright assignment
are primarily those who already have reason to believe their employers
will be amenable to it.  It's difficult to measure the number of
developers who publish free software who *might* contribute directly
but believe their employers will not agree, or simply do not want to
involve their employers in their private hobbies, much less the number
of people who are wholly deterred from even writing software that
might be published under a free license because their employer might
simply dislike their doing so, whether the employer has an enforceable
right to .that copyright or simply intends to demand it regardless of
the enforceability of the claim.  In practice, the purpose of such
expansive claims to work produced outside the course of employment is
primarily to discourage any productive effort outside of that done for
the employer.

> That doesn't seem to be relevant to the clojure-mode situation,
> we now know.  The developers objections are others.

I'm still not clear to me what the objections really are, if the labor
of obtaining the CAs and adapting a version to the emacs development
process were provided.  This is assuming that the GNU emacs project
won't privilege an external project for claiming a standard naming
pattern such as "<language>-mode" with no intention of contributing it
to the project, and will simply use the name "clojure-mode" as it
deems appropriate.  It doesn't seem the current situation was
anticipated.

Lynn