[License-review] Approval Request: Free Public License 1.0.0

[Rick Moen]

Quoting Christian Bundy (christianbundy at fraction.io):

> The FPL is a modified ISC license that removes both the copyright
> notice and the requirement that "...the above copyright notice and
> this permission notice appear in all copies." While there's only a
> small textual difference between the two licenses, removing the
> copyright notice and copyright requirement solve the problem of
> software public domain dedication.

I see this as an intelligent and respect-worthy expression of the
recurring effort to create a permissive licence that's absolutely 
as close as humanly possible to nullifying copyright, within the
Berne Convention regime we all must contend with.  (If any part
of that in inaccurate, apologies, and I seriously intend no criticism or
offence in the above characterisation.)

Nothing in FPL 1.0.0 strikes me as OSD-problematic, at a quick glance.
And I'll even call it innovative, to my eyes.


Nonetheless, I'd like to register my view that a key aim and component
of your licence is very unwise: the omission of copyright notice,
omission of the requirement to retain the permission notice, and omission 
of the requirement to retain the copyright notice in all copies.
Because copyright as a legal reality cannot be finessed away, code
recipients benefit greatly from knowing what copyright encumbrance
exists and what year it arose (thus, what year it will cease to exist),
and what licence terms the stakeholder specified.  

How can I best get this across?  The good old days when a large fraction
of code tarballs on the Internet's ftp sites omitted all mention of 
copyright names / dates and licensing:  Those were _bad_ days.  Pulling
down such a tarball and just hoping your derivative works wouldn't lure
some stakeholder out of the weeds to take your lunch over a copyright
tort, being obliged to treat all such works as under a cloud of possible 
serious problems, we should not want to go back there.

The _issuer_ of a FPL 1.0.0 would have no problems.  He or she knows that
it's genuinely permissive-licensed.  It's merely the poor sods who
_receive_ copies, particularly at greater remove from the issuer, who
are left in a problematic information void.

You should not want third-party code reusers to have that problem, and I
certainly don't.

You're more than welcome to believe that obligatory copyright notices,
and obligatory permission notices, are an impermissible infringement on
total software freedom.  I respect that view.  However, works
distributed that way are, as provided, and unless the copyright world
changes, indistinguishable from works made available in violation of a
(unstate) person's copyright.  The uncertainty thus created is a lurking
menace to third parties.

The above are, honestly, addressed primarily to OSI itself, because 
FPL 1.0.0 strikes me as well drafted to achieve a specific effect,
and so conclude that you at Fraction know what you are doing.  If I felt
otherwise, I'd be trying to persuade you it's a bad idea.  Instead, 
I'll just say to OSI that I hope it will feel no desire to approve 
a permissive licence that creates this class of problem. 

If it were up to me, I'd also just not waste OSI's time on new licences
'more permissive than traditional licenses like the MIT license',
because anything less than that is frankly problematic.  (However, in no
way do I speak for OSI.)

-- 
Cheers,                   You can't make anyone see reason.  All you can do is 
Rick Moen                 throw a big party around reason and see who shows up.
rick at linuxmafia.com                                          -- @alexandraerin
McQ! (4x80)