[License-review] For Approval: License Zero Reciprocal Public License

[Kyle Mitchell]

Nigel,

Thanks again for time and attention.

I think you're generally correct about how users
get licenses.  With some exceptions, notably The
MIT License, popular Open Source licenses I'm
aware of don't grant rights to sublicense.
Rather, they require preservation of the license
terms with copies, so that folks downstream become
aware that they have permission direct from
licensors, under the public license.  Heather
Meeker calls this "direct licensing" in her books.

I'm not sure you're correct about the effect
direct licensing has on compliance with license
conditions.  Users receive licenses directly from
rights holders, but the terms of those licenses
are the terms of the applied public licenses,
conditions and all.  Whether those conditions
actually have any effect on folks we're calling
"users" depends on the specific wording of the
condition.  Are the conditions triggered by things
that users do?

Terms vary.  For example, Apache-2.0's
"Redistribution" section actually speaks in terms
of "reproduc[ing] and distribut[ing] copies", but
all the requirements apply either to changes made
(b) or distribution (to "recipients") (a, c, d).
Conversely, MIT's attribution condition speaks
only in terms of "copies".  I could argue that
language reads on private copies, too.

Your concern about affecting other code due to
even casual use is a very valid one, and a common
refrain in discussions of copyleft terms.  L0-R
has even more of a hair trigger than AGPL, by
design.  That fact directly motivated the
built-in, n-day grace period.

The intent is _not_ to turn well meaning folks
into unwitting infringers.  Rather, it's to
motivate contributions to what's available under
Open Source terms.  Contributions may come either
in kind, as released code, or in coin for
alternative license terms, to support those
releasing code.

Best,

K

-- 
Kyle Mitchell, attorney // Oakland // (510) 712 - 0933