[Python-Dev] Still no new license -- but draft text available

[Tim Peters]

[Gary Momarison]
> Tim, I'm glad to have had a response from you,

It's not like an audience with the Pope or anything, but I can understand if
it's a thrill <wink>.

> who I have found (in reading the last few days of posts) to be a
> good thinker and writer. (I'm new to the newsgroup).

Stick around!  It's usually a lot more fun than this -- Python *had* lived a
charmed life until recently, from picky legal points of view.

> I'm sorry you didn't feel qualified to address my main issue
> more directly, but I hope the exchange will help me and others who
> will need to be involved in composing or selecting a license
> in the future.

Ah, that's a different matter entirely.  The situation wrt the Python
license is mostly in CNRI's hands, but if you're working without constraints
the cardinal rule is "don't innovate!".  That is, pick the existing licenses
you like best from the list at

     http://www.opensource.org/licenses/

and ask an attorney which suits your goals best.  The GPL is a great choice
if your project can live with its implications, and you're comfortable
assigning copyright to the FSF.  I happen to think public domain is severely
underused these days too (but then you give up copyright entirely).  Guido
is fond of the BSD license.

Courts have been known to do surprising things, and unless you view yourself
as an Intellectual Property Pioneer, I think you're better off picking a
license that scads of other people rely on too (if nothing else, then if
you're attacked, they'll feel at risk too, and at least America is an "all
the justice you can afford" kind of country -- unless you've got deep
pockets, you'll go broke defending a unique license on your own).

As to the rest, I have to repeat that I'm flatly incompetent to chop legal
technicalities.  Not only not a lawyer, but have no native interest in the
subject.  Just two glosses:

> ...
> My point was just that RMS's "ruling" on this license is irrelevant
> except to the extent that his well-informed but non-lawyer's opinion
> influences potential licensors and licensees.  You probably should
> care about that to popularize Python, but it has nothing to do with
> the validity of the license in court.  RMS's "ruling" won't be a
> factor in court.

"RMS says XYZ" is actually shorthand for "RMS has consulted with FSF's legal
counsel, and the FSF's position is XYZ".  The FSF has had the active help of
attorneys and law professors for about 20 years, and RMS doesn't hesitate to
ask if there's any point about which he's uncertain.  So far this appears to
have kept the GPL out of court entirely ("the other side" has always backed
down first).

> ...
> I'd hate to see everyone just give up on freer-than-GPL licenses
> just because they are confusing or poorly written from a legal/
> corporate perspective.  Let's hope the new Python license is an
> improvement or at least contains some stuff that will lead to
> improvements.

In recent times, the Open Source Initiative has been saying there are
already plenty of licenses to choose from.  Since CNRI was determined to
roll their own, we're especially grateful that the board of the OSI was
willing to examine it.  Visit

    http://www.opensource.org/

and check it out.  Discussion of new Open Source licenses is carried out on
their license-discuss mailing list, which you can subscribe to via sending
email to

    mailto:license-discuss-subscribe at opensource.org

comp.lang.python isn't a particularly good place for this, as the new
license is  CNRI's and they're not discussing it here.  Of course, they
would be more than welcome to!  Whether people can actually use Python 2.0
at all may be almost as important as whether list comprehensions make it in
<0.3 wink>.

y'know-most-people-ignore-all-this-and-most-people-never-have-
    cause-to-regret-it-ly y'rs  - tim