[conspire] (forw) Re: Can OSI specify that public domain is open source?

[Rick Moen]

As I expected, just another affronted author of a one-off, badly written
permissive licence unwilling to comprehend and deal with being ignored.


----- Forwarded message from Rick Moen <rick at linuxmafia.com> -----

Date: Thu, 8 Sep 2011 16:35:47 -0700
From: Rick Moen <rick at linuxmafia.com>
To: Chad Perrin <perrin at apotheon.com>
Subject: Re: Can OSI specify that public domain is open source?
Organization: If you lived here, you'd be $HOME already.

Quoting Chad Perrin (perrin at apotheon.com):

> There are a couple of weird edge cases where advertising clauses disagree
> with each other or something along those lines....

Yeah, good point.

> Gaining popularity is difficult, to be sure.  Look at the long slog OSes
> like the various BSD Unix systems have had to gain (or sometimes even
> maintain) any kind of popular recognition, especially by comparison with
> MS Windows.

Popularity of a codebase is a bit of a different problem space from
the popularity of a licence.  Once a licence is seen as useful and well
known, there can be no technological friction slowing down its adoption.
Witness for example the very widespread use of 2-clause BSD and its
equivalent the MIT Licence in applications neither the UC Berkeley
Computer Science Research Group nor the MIT X Consortium could have even
conceived of.

The fact that we now can spot some significant legal weaknesses in those
licences -- patent defence, legal assent -- is powerfully outweighed by
the fact that they're court-tested, extremely clear, and extremely
well known.


> If nothing else, the OWL has a place in category two, where there's
> nothing in your list approaching the kind of simplicity of language and
> effect it offers.

No, you may feel it _deserves_ a place, but it very doesn't have one and
almost certainly never will, because (1) it's not well drafted, (2) it's
far too late to the party, (3) there's no compelling interest behind it
(shouldn't be a factor in a fair, just world with no transaction
friction in evaluating information, but this is not a perfect world),
and (4) it provides nothing that's compelling.

Obviously, you differ with me on (at minimum) consideration #4, but
that's to be expected because you're the author.  You say 'simplicity of
language', but I find MIT Licence so ridiculously easy -- for,
obviously, software application only -- to read that wasting everyone's
time writing something simpler cannot be justified.  In my opinion;
you're welcome to your different one.

By the way, if I wanted to use a permissive licence for a non-software
work and wanted MIT Licence's extreme simplicity, I'd do:
Copyright (C) 2011 by Rick Moen <rick at linuxmafia.com>.  This work may
be used subject to the terms of MIT License despite not being software.

(It is good form to also include the 17 lines of the licence itself, but
not strictly necessary.)



[AFL:]

> Unfortunately, I think part of its popular failure was the fact that it's
> lengthy and wordy -- and people are unlikely to read it all the way
> through even once.

Their loss.  All of that is there for a reason.  And I get tired of
non-lawyers trying to say that people like Larry Rosen don't know what
wording is necessary and desirable.  Good luck with that, folks.

You know, I'm sorry, but this whole thing is extremely reminiscent of
the complaint we in the Unix community keep hearing from technophobes:
'The software is too complex.'  Sorry, some things are irreducibly
complex because there are necessary parts that cannot be safely omitted.
You can ignore that, of course.  The holes you shoot in your own and
others' feet are your own (and their) problem.



> While the GPLv2 does not mention specific application only to software
> (as far as I recall -- I haven't read it in a couple years), it
> definitely imposes restrictions and conditions that make no sense for the
> majority of content forms.

As always, the licence should reflect the licensor's intent (modulo
other concerns, such as licensor wishing to not render his/her work
irrelevant through terms that will be perceived as gratuitously
different).  For many of us, the restrictions and conditions GPLv2 
imposes are exactly those we intend, e.g., we like the idea of
commercial redistributors being obliged to offer the preferred form for
three years.  You don't care about that, fine, but don't say it makes no
sense for the majority of content forms.


> Some of its conditions might have bizarre and undesirable effects for
> some types of work, too.  For instance, one might make an argument
> that applied to music it would prohibit you from sharing a recording
> without also offering the sheet music to the recipient.

I do not buy the crack-addled notion that the preferred form of an MP3
is sheet music, sorry.



> "You may not impose any effective technological measures on the Work
> that restrict the ability of a recipient of the Work from You to
> exercise the rights granted to that recipient under the terms of the
> License."

CC drafters think Tivoisation sucks.  If you don't object to
redistributors imposing DRM handcuffs to control recipients, waive that
clause.  Costs you about five words, right below your copyright notice.


[my two-line ultra-permissive licence:]

> You might want to rephrase that.  It seems to indicate that a
> redistributor cannot offer a warranty of his own to subsequent
> recipients.

No, it doesn't.  As usual, context matters.  The indicated phrase is
_George Coder_ speaking.  Obviously, he is entitled to speak, and is
only speaking, for himself.  If you showed up in court and argued that
George had some magical power to bind independent creators of derivative
works and deprive them of the lawful power to offer warranties on their
own products, the judge would bounce you down the courthouse steps.


[CC licences:]

> They're also widely misunderstood, tend to have legal landmines embedded
> in them, and difficult to read through without falling asleep.

'Legal landmines' my great aunt.  Understanding them is not the least
bit difficult for anyone at a teenage reading level, _and_ the
integrity of CC in making sure the 'plain English' version of the
licence matches the 'legal code' version is well known to be
trustworthy.  So, if for some reason you're not capable of reading the
latter, you can use the former as Cliff's Notes.


> I guess they're "competent" in the eyes of a lawyer....

And judges.  Who do you think applies them?

> but in actual practice....

In actual practice when applied by... judges.  Thus my point.  You write
a bad licence -- such as the one you've already written -- and you're 
playing Russian Roulette with how it will eventually be interpreted by a
judge.  That is, if a bloody miracle happens and your badly written
licence actually becomes significant.  Fortunately for random users of
creative works, badly written permissive licences get ignored.  Just as
yours will.

> ...often cause harm by way of the effect they have on the people who
> have to actually apply licenses to their works.

Bullshit.  It's really simple, Chad.  You read a few licences (and, if
you're too much of a lightweight for the CC 'legal code', stick to the
'plain English' variants.  You pick whichever one best matches your
intentions.  If you have any common sense, you pick one that won't
hinder your work's use by its contemplated userbase by motivating them
to say 'What's this squirrely licence?  I'll leave it for later.'

> Oddly, it seems like your recommendation of these licenses flies in the
> face of your own complaints about proliferation of licenses that are not
> popular enough, and probably never will be popular enough.

Then, you didn't frelling _bother_ to read what I said.

I said that AFL and OSL are very meritoriously written, but arrived
late, and other licences are already court-tested, clear, and widely
understood, and therefore are doomed despite their intrinsic merit.


You know, Chad, I cannot justify continuing this private conversation,
and it is futile.  If you want to do yourself a favour, read Larry's
book.


----- End forwarded message -----