Re: Squeak in Debian?
Martin, it's great of you to do a summary. My thoughts included below.
Henning Makholm <henning@makholm.net> wrote:
> Scripsit Martin Schulze <joey@infodrom.org>
>
> > | "You may distribute and sublicense such Modified Software only under the
> > | terms of a valid, binding license that makes no representations or
> > | warranties on behalf of Apple, and is no less protective of Apple and
> > | Apple's rights than this License."
> > |
> > | What the heck does that even *mean*? Licenses aren't "binding"; they're
> > | thinking of contracts. In fact, the whole license thinks it's a contract
> > | (which is bad from the start). "Protective of Apple and Apple's rights" is
> > | incredible vague, meaning that only this exact license is a safe license
> > | for derivative works.
>
> > What's this?
>
> Insufficient data (as quoted: What the heck does this mean?). In the
> absence of better information I'd prefer to err on the side of caution:
>
> [X] Renders the package non-distributable
I do not see a problem with either issue here and think that whoever
posted it was simply reading too quickly. The text seems clear to me
in all areas that matter, even though I do not know what "equally
protective" means, exactly, either.
First, licenses that count *are* binding. Squeak-L is preventing you
from shipping someone a compliant license while the real terms are under
some other license; the restrictions on sublicensing Squeak-L apply to
the *real* license, the license that the receiver is going to be bound
by, not by any other license that you might have concocted. Squeak-L is
just closing potential loopholes by tossing in the "valid" and
"binding".
Second, the "equally protective" aspect is giving distributors an
*extra* permission that we do not plan to use. There's no need for us
to distribute Squeak under anything but Squeak-L, and Squeak-L is surely
"equally protective" of Apple as Squeak-L.
> > | This is a forced-distribution clause. It requires that if the Modified
> > | Software is given to *anyone*, it must be made "publicly available" (to
> > | lots of *other* people) at no charge. This fails the dissident test and
> > | the desert island test. It's also a practical inconvenience; I can't share
> > | a modified version with my spouse without publishing it.
>
> > What's this?
>
> [X] Renders the package non-free
I disagree, as posted earlier. It's not important, though, because the
export regs already likely make the software not be DFSG-free.
> > | "This License allows you to copy, install and use the Apple Software on an
> > | unlimited number of computers under your direct control."
>
> > | Purports to restrict use. Doesn't allow use on computers not "under your
> > | direct control", which is a substantial restriction; it probably prohibits
> > | it from being installed by a Debian admin onto a Debian machine which is
> > | hosted elsewhere. :-P
>
> > What's this?
>
> Confused drafting, probably leftover from adaptations of commercial
> licenses that restricts the number of installations.
>
> I'm not convinced that this is necessarily a DFSG problem, but if one
> has to get clarification from upstream about the "valid, binding
> license" part above, one might just as well ask for this to be
> clarified (or left out entirely).
I read this as Squeak-L drawing the line between usage and distribution.
If you cross the line then you do not get to use the "usage"
permissions, but you can still use the "distribution" permissions.
There may be a free-ness issue, in that Squeak-L may draw the line more
narrowly than is assumed by DFSG; thus some DFSG "uses" would be
Squeak-L "distributions", and Squeak-L distribution permissions may be
insufficient for DFSG's requirements for usage.
It's irrelevant, though, since Squeak-L is DFSG non-free anyway. The
open question is whether we can distribute the software safely. For
answering that question, we clearly need to look at the distribution
provisions.
-Lex
Reply to: