Date: Tue, 14 Nov 2017 10:10:07 -0800
From: "Kyle E. Mitchell" <kyle@kemitchell.com>
To: License submissions for OSI review <license-review@opensource.org>
Subject: Re: [License-review] For Approval: Rewrite of License Zero
 Reciprocal Public License
Message-ID: <20171114181007.GL15549@dev.kemitchell.com>
References: <20171027202935.GA5874@dev.kemitchell.com>
 <CAK2MWOta_sCbfiztv=yfT_m6xBNnMqWPUy=iZukFWmypb-UdzA@mail.gmail.com>
 <20171107224929.GK25059@dev.kemitchell.com>
 <CAK2MWOvhJCMFimdGSV5xw2MF8CzfjZt1DKJZKXh1GGeKMngiTQ@mail.gmail.com>
 <20171108075050.GQ25059@dev.kemitchell.com>
 <CAK2MWOvXOQNvRsNXyQd06o0jw9g0fc=GB4zYJVYsaHzbW+fOdg@mail.gmail.com>
 <CAD2gp_QE_0dz7m3nE8uMECMSYkXD64FP=r6s4LoHx_n7obQNCQ@mail.gmail.com>
 <CAK2MWOt2X8XHfUEtrDp4Tzu9vGe2fD9+ydV4BUir08CgnTEksQ@mail.gmail.com>
 <34f8ca5c-e16b-51a7-ce0f-b2ec88714a1f@piana.eu>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <34f8ca5c-e16b-51a7-ce0f-b2ec88714a1f@piana.eu>
User-Agent: Mutt/1.5.23 (2014-03-12)

On 2017-11-08 19:11, Carlo Piana wrote:
> On 08/11/2017 18:51, Bruce Perens wrote:
> > On Wed, Nov 8, 2017 at 9:34 AM, John Cowan <cowan@ccil.org
> > <mailto:cowan@ccil.org>> wrote:
> >     On Wed, Nov 8, 2017 at 11:27 AM, Bruce Perens <bruce@perens.com
> >     <mailto:bruce@perens.com>> wrote:
> >         In your license, you are asking for an *unrelated* program to
> >         be made Open Source due to a condition predicated on a
> >         specific form of use. Besides being clearly against OSD # 6
> >         (and sorry, OSI is not now required to announce this obvious
> >         fact),
> >
> >     I don't see this.  If OSD #6 means, or is interpreted to mean, "no
> >     discrimination against particular tasks", then I can see it, and
> >     that's a license prohibition I would support.  But I have trouble
> >     reading it into the text of #6.  Can you explicate?
> >
> > Here is the relevant L0-R text:
> >
> > /5. If you run this software to analyze, modify, or generate
> >      software, you must release source code for that software./
> > /
> > /
> > Analysis, modification, and generation of software are a field of
> > endeavor under OSD # 6. OSD #2 says the /program /must include source
> > code, so it overrides any provision in OSD # 6 that would prevent the
> > program from having source code. But not /other /programs. So, GPL OK,
> > King Midas not OK.
> >
> > We strengthen this with OSD # 9 requiring that the license not bind to
> > other software that is simply distributed with the Open Source.
>
> [...]
>
> Il love this King Midas analogy. It is quite graphic and I could not
> agree more with Bruce. The provision Bruce has extracted is not an
> acceptable one in open source. I have already stated my case, many
> others have, I stand firm in my belief. I am shocked we are still
> arguing something that should have been long settled by now.

Midas starved because his touch turned his food into gold he
couldn't eat. I recall a version where he turns a daughter
to gold, too. Very sad.

That's what proprietary software does to permissively
licensed code. Building on permissively licensed code makes
a proprietary program better, and therefore more marketable.
The open components transform from raw materials into gold.
But the proprietary touch also hardens them, locks them up
behind confidentiality and restrictive terms like bullion in
a vault. That prevents anyone from further growing and
developing them in application, where they matter most. So
the Open Source involved becomes undigestable for users and
the developer community, impossible to break down and turn
into what's needed.

Companies don't starve for lack of food. Gold is what they
need to stay alive. But still their offspring---products and
services---could often bring more gold, long term, if they
stayed open, and developed.

Reciprocal licenses are the grace Midas longed for when he
tasted gold in his mouth. They encourage turning proprietary
code, petrified for the sake of immediate merchantability,
into code that's malleable, that lives again, that can grow
in intrinsic value over time. Their touch may be extensive,
but what they touch turns into open source. Not sad at all!

"Reciprocal" is a great name for these licenses, if I don't
say so myself, because give and take is what they're about.
There's a deal on offer---use my code, but share yours
back---that's very appealing, but nobody has to accept it.
No proprietary project is "infected" with reciprocal
licensing any more than Ebenezer Scrooge is "infected" with
Christmas, another great mutual-giving tradition.

> Our world has fought long battles against the characterization of
> copyleft as "viral", which in turn is intended in the pejorative
> "infectious", a frivolous yet still today damaging concept that detracts
> many from using copyleft software. A license containing such language
> would be indeed flat out infectious, the legal equivalent of a virus.
> The fact that this would also be legally sound, just worsens the
> situation. Let's imagine a Patrick McHardy using that clause. You must
> be kidding me.

I'm so grateful to those who fought anti-copyleft FUD in its
heyday. Many on this list were directly involved, I know.
That nonsense ought to stay in its grave, where you put it.
And the community is freer to advance since you did. Heck, a
principal corporate proponent of that rhetoric is recently a
member of OSI!

FUD like "viral" wasn't wrong because it exaggerated how
strong reciprocal licenses were at the time. It won't become
true as reciprocal licenses get stronger. Viral license FUD
is, was, and always will be basically, fundamentally,
intentionally false. A deception.

Courts do not issue orders to comply with privately drafted
license conditions. No court applying law I know, hearing a
lawsuit about GPL or L0-R software, would ever issue an
order to release and publicly license proprietary source
code. If they find a licensee exceeded their permission by
failing to meet conditions, with no legal excuse or defense,
they'll order payment of money damages, possibly attorneys'
fees, and perhaps issue an order to stop using the software.
I believe that's it.

I've stopped posting revisions of L0-R's license text so
frequently, because that proved hard to follow on the list.
But here's my current automatic-forgiveness language, akin
to GPLv3 section 8:

  Any unknowing failure to meet [the reciprocity conditions]
  is excused if you release source code as required, or stop
  doing anything requiring permission under this license, in
  30 days of learning that you were required to release.

If Patrick McHardy sends you a threat letter about L0-R code
you didn't know you were infringing, you have 30 days to
stop using his code, grep his name in your codebases, and
put a policy in place to keep it out going forward. Then you
get to write him a letter saying he can go pound sand. Send
me a copy, and I'll post it on my website, frame a copy, and
hang it in my office.

That's how I read the language above. If that's not how it
reads to you, let me know. I want to fix it.

---

I have great respect for the voices and contributions of
those arguing against conformance. But I don't understand
your positions. Calls that it's obvious, or that the
contrary is absurd, don't help anyone who doesn't already
agree with you to understand. They block consensus, sure.
But if consensus is the goal, in one direction or another,
blocks are cause to talk more, not less.

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