New OSI-approved licenses

Rob Landley rob at landley.net
Sat Dec 5 01:37:55 UTC 2015


Did this ever get resolved?

On 11/17/2015 12:51 AM, Richard Fontana wrote:
> On Tue, Nov 17, 2015 at 01:32:44AM -0500, Richard Fontana wrote:
>>>> 2) Free Public License 1.0.0
>>>> Text of approved license contained within:
>>>>
https://lists.opensource.org/pipermail/license-review/2015-August/001104.html
>>>
>>>  We have added as of v2.2 - http://spdx.org/licenses/0BSD.html -
although it was submitted using a different name as suggested by the
submitter (who I think said he authored the license… or at least seemed
to know a lot about it’s origins and the suggested name, which we went
with - see
http://lists.spdx.org/pipermail/spdx-legal/2015-June/001443.html for
that thread).
>>>
>>> Would the OSI oppose the name we already went with??
>>
>> Hmm, I think this is really a case of two people independently
>> inventing approximately the same thing at about the same time,

Given the timeline, it's more likely they copied the license from Android.

March 2013:

I start using this license:
http://lists.landley.net/pipermail/toybox-landley.net/2013-March/000794.html

November 2014:

Android merges toybox to replace toolbox:
https://code.google.com/p/android/issues/detail?id=76861#c11

January 2015:

Linux Weekly News covers toybox's addition to Android:
https://lwn.net/Articles/629362/

May 2015:

Android-M preview containing toybox distributed to developers:
https://en.wikipedia.org/wiki/Android_Marshmallow

June 2015:

Either Samsung or Sony (I forget which) asks me to submit the the toybox
license to SPDX to simplify their internal paperwork:
http://lists.spdx.org/pipermail/spdx-legal/2015-June/001443.html

August 30, 2015:

These guys submit the license to OSI.
https://lists.opensource.org/pipermail/license-review/2015-August/001104.html

I didn't submit my public domain equivalent license to OSI because their
lawyer wrote an article literally comparing public domain software to
abandoning trash by the side of a highway
(http://www.linuxjournal.com/article/6225 paragraph 5), and if you
google for "Linux Public Domain" it's still the second hit.

>> where
>> 'invention' means removing some language from an existing short
>> license.

Yes, it is a minor variant of an existing BSD license.

Specifically, the OpenBSD template license
(http://www.openbsd.org/policy.html links to
http://cvsweb.openbsd.org/cgi-bin/cvsweb/src/share/misc/license.template?rev=HEAD)
which is why I felt justified in calling it a BSD license.

There were already "2 clause", "3 clause", and "4 clause" BSD licenses
commonly referred to, "zero clause" to mean public domain didn't seem
like a stretch (and is also what the Creative Commons guys chose with
CC0). It also makes it easy for corporate legal departments that have
approved existing BSD licenses to rubber-stamp another.

I chose this name for a reason. In the "copyleft vs bsd" axis, this is
more BSD than BSD. "Free" is the Free Software Foundation's rallying cry
(and the reason OSI had to come up with "Open Source" to counter "Free
Software"). Sticking the word "Free" on a public domain equivalent
license (as far from copyleft as you can get) is either intentionally
confusing or deeply clueless.

Part of my attraction to public domain licensing is trying to counteract
the damage GPLv3 did to the community at large when it fragmented
copyleft into incompatible factions. There's no such thing as "The GPL"
anymore, Linux and Samba implement two ends of the same protocol, are
both GPL, and neither can use the other's code. Copyleft is now a
significant _barrier_ to code reuse within copylefted projects.

The result seems to be a generation of programmers who are lumping
software copyrights in with software patents as "too dumb to live", and
taking a napster-style civil disobedience approach, opting out of
licensing their code at ALL until the whole corrupt intellectual
property edifice collapses under its own weight. "No License Specified"
continues to be the most common license on github _after_ its CEO made a
big push to standardize on MIT licensing as a default. The percentage
has gone _up_ in the past year:

https://speakerdeck.com/benbalter/open-source-licensing-by-the-numbers?slide=41

If I just wanted a public domain license I could have grabbed creative
commons zero (or the libtomcrypt license or unlicense.org or...) but I
wanted the strategic advantage of the name "Zero Clause BSD" because the
ability to say "we're more BSD than BSD" is an easy sell that
short-circuits a lot of explanation.

Attaching the Free Software Foundation's codeword "Free" to a
non-copyleft license is... odd. Saying "when we use the word 'Free' we
mean something different than when the Free Software Foundation uses the
word 'Free'" is not an argument I want to make, especially not to people
who have developed an _aversion_ to copyleft.

> Looks like Rob Landley was using it a year or more earlier:
> https://lwn.net/Articles/608082/

According to http://landley.net/hg/toybox/rev/264b9da809df since March
14, 2013, so 2 and a half years earlier than August 30 of this year.

Checking my old email, I noticed 0BSD marked approved for 2.2 in your
public spreadsheet July 19, 2015, so you'd already approved it a month
and a half before the other guy submitted it to OSI.

I'm guessing what happened here is that Android's "about->licenses"
thing gives license text but not license names, so they made one up.

Rob


More information about the Spdx-legal mailing list