License recommendation for specs

[Simon McVittie]

On Wed, 02 Jan 2019 at 14:02:12 +0100, Egmont Koblinger wrote:
> For specifications, even for "free" ones, I think it's outright undesirable to
> create and distribute modified work. In order to avoid incompatible
> specifications competing against each other, resulting in quite a mess and poor
> interoperability between implementations, contributors should be pushed to
> improve the original specification whenever feasible

I agree that incompatible specifications are undesirable, but that
doesn't mean that copyright law is necessarily an appropriate tool
to prevent incompatible specifications. If I write a specification
for freedesktop.org, and 20 years later you are among freedesktop.org
contributors wanting to base a new interoperable specification on the
one that I started, it's likely to be practically problematic to expect
you to track down my contact details and get my permission.

Copyright says the author (or their employer) controls copying and
derivative works, but specifications should usually be controlled by
some sort of community consensus process.

> Not sure if this restriction belongs to the license, though, or should sort
> itself out by other means, e.g. by the original work having some "respect" and
> the community refusing to go with forks.

I think this is a better route to take.

If you insist on having the unique canonical version of the specification
backed up by law, trademarks are probably a more appropriate tool than
copyright.

> Also note that even if the license chosen for a specification disallows the
> creation of modified work, it should probably still be allowed if done with the
> explicit intent of sending suggested modifications back to mainstream. (If I
> understand correctly, CC BY-ND doesn't allow the creation of publicly visible
> merge request?)

Correct. CC-BY-ND does not allow derivative works, however helpful the
intention of those derivative works might be.

> Debian's guideline of being "free" instead seems to focus on the modification
> and distribution of the material in question instead.

The Debian Free Software Guidelines are applied equally to all software,
including non-code material such as documentation and specifications.
We consider a specification to be Free Software if we can copy it, modify
it and so on.

It's common to want to quote particularly subtle or important sections of
a specification in an implementation of that specification (for example in
a comment in C code), which results in the implementation being a derived
work of the specification. If the specification is under a copyleft,
restrictive or non-Free license, that's problematic.

> My instincts tell me that a "free specification" should be one that everybody
> is free to implement, without imposing restrictions on the implementation.

Older Internet RFCs are a prominent example of specifications that are
definitely free to implement (in software under a wide range of licenses),
but are under licenses that do not allow derivative works. (As a result
you can't quote particularly tricky pieces of older RFCs in your code
comments, unless allowed by fair use or your jurisdiction's equivalent
concept.)

> it's unclear whether an implementation of a specification counts as "derived
> work" or not

This is a question of copyright law, not licensing. Licenses like the
GPL and CC-* delegate the question of "what is a derivative work?" to
the relevant jurisdiction's copyright laws. A copyright license like the
GPL or CC-* is permission from the copyright holder to do some limited
set of things that would otherwise be copyright-infringing; it cannot
prevent actions that would have been allowed by copyright law anyway.

> Ideally I
> wouldn't want others to distribute my work (the specification itself, or parts
> of it) commercially

If a license doesn't allow commercial redistribution then it isn't
Free Software (don't confuse "non-commercial use only" clauses with
copyleft).

    "A free program must be available for commercial use, commercial
    development, and commercial distribution."
    — https://www.gnu.org/philosophy/free-sw.en.html

The Debian Free Software Guidelines and the Open Source Definition also
require commercial use and redistribution to be allowed (in case you aren't
aware of their history, the OSD is heavily based on the DFSG).

> Or, what happens if I just release without any licensing? What are the
> downsides of that approach? Why is it important or encouraged to have a
> license? Aren't the "legal defaults" of any published work without a license
> good enough, and perhaps closer to what I imagine than any of these existing
> license texts?

No. The default, in the absence of any copyright license, is that nobody
is allowed to copy and redistribute your specification (copyright law
does not distinguish between a specification and the latest Disney film).

    smcv