From Fernando.Perez at colorado.edu  Fri Oct  1 14:14:06 2004
From: Fernando.Perez at colorado.edu (Fernando Perez)
Date: Fri, 01 Oct 2004 12:14:06 -0600
Subject: [IPython-dev] IPython license switch: LGPL -> BSD
Message-ID: <415D9E6E.4020604@colorado.edu>

Hi all,

if you are getting this email and you are NOT on the ipython user/dev lists, 
it is because I have your name on the credits list for IPython 
(http://ipython.scipy.org).  This means you may have contributed code, ideas, 
bugfixes, or some other form of help to iypthon in the past.  I'm not trying 
to spam you, really :)

I'd like to let everybody know that for the next release (0.6.4, most likely) 
I plan on switching licenses for ipython from LGPL to BSD.  The reason behind 
this switch is to enable smoother integration with Scipy 
(http://www.scipy.org) and matplotlib (http://matplotlib.sourceforge.net).

In order to spread further the use of python for scientific use, we want to 
provide a set of tools which is as well integrated and easy to use as possible 
for scientists who are not necessarily python experts.  Scipy is BSD[1] 
licensed and matplotlib uses a PSF-type license[2], but up to this point 
ipython has been LGPL[3], so it makes code integration a problem.  I wish to 
change this.

I have already contacted the other two core authors of ipython's code, and 
they are OK with the switch.  Additionally, ipython distributes (unmodified) 
Bill Bumgarner's DPyGetOpt and Ka-Ping Yee's Itpl, but these modules are 
MIT[4] licensed, so there is no problem there either.  I will leave those 
modules as they are today.

I am giving this public heads up in case any of you has contributed _code_ to 
ipython in the past, and disagrees with this license change.  I would regret 
such a situation, and resolving it would be very time-consuming, as I haven't 
tracked individual patches with names carefully.  So I'd have to dig through 
CVS quite a bit to remove and rewrite third-party contributions which don't 
want to be part of a BSD-released IPython.

But if anyone has very strong objections on this matter, please point me to 
the code you've sent me in the past, and I'll do my best to excise it from 
ipython and rewrite the necessary functionality myself from scratch.

Please note that I am NOT soliciting feedback on this decision, nor am I 
proposing a discussion on licenses.  This decision is made and final.  So 
unless you:

1. have sent me code for ipython in the past, AND

2. object to such code being distributed under a BSD license,

you can safely ignore this message.

I will wait for 2 weeks from today, and if I hear no replies otherwise I'll 
consider all current code in ipython OK for BSD relicensing (except for the 
above mentioned MIT-licensed code, which will remain unchanged).

Best regards to all, and sorry for the unsolicited message.

Fernando.


REFERENCES

[1] http://www.opensource.org/licenses/bsd-license.php, see also 
http://www.scipy.org/About/FAQ.html#license

[2] http://matplotlib.sourceforge.net/license.html

[3] http://www.opensource.org/licenses/lgpl-license.php, see also 
http://www.gnu.org/copyleft/lesser.html

[4] http://www.opensource.org/licenses/mit-license.php



From vivainio at kolumbus.fi  Sat Oct  2 09:17:40 2004
From: vivainio at kolumbus.fi (Ville Vainio)
Date: Sat, 02 Oct 2004 16:17:40 +0300
Subject: [IPython-dev] IPython license switch: LGPL -> BSD
In-Reply-To: <415D9E6E.4020604@colorado.edu>
References: <415D9E6E.4020604@colorado.edu>
Message-ID: <415EAA74.4050504@kolumbus.fi>

Fernando Perez wrote:


> the switch.  Additionally, ipython distributes (unmodified) Bill 
> Bumgarner's DPyGetOpt and Ka-Ping Yee's Itpl, but these modules are 
> MIT[4] licensed, so there is no problem there either.  I will leave 
> those modules as they are today.


Why not use the MIT license for the whole ipython? On a c.l.py 
discussion that took place a while ago, the MIT license was seen as more 
straightforward and "obviously free" (nothing is free enough to some 
die-hards) than the BSD license.

Or is:
"""
Neither the name of the <ORGANIZATION> nor the names of its contributors 
may be used to endorse or promote products derived from this software 
without specific prior written permission.
"""

important to you in some specific manner?


> Please note that I am NOT soliciting feedback on this decision, nor am 
> I proposing a discussion on licenses.  This decision is made and 
> final.  So unless you:


Oops, I did it already :-). BSD is fine with me.



From Fernando.Perez at colorado.edu  Sat Oct  2 15:28:29 2004
From: Fernando.Perez at colorado.edu (Fernando Perez)
Date: Sat, 02 Oct 2004 13:28:29 -0600
Subject: [IPython-dev] IPython license switch: LGPL -> BSD
In-Reply-To: <415EAA74.4050504@kolumbus.fi>
References: <415D9E6E.4020604@colorado.edu> <415EAA74.4050504@kolumbus.fi>
Message-ID: <415F015D.4000201@colorado.edu>

Ville Vainio wrote:

> Why not use the MIT license for the whole ipython? On a c.l.py 
> discussion that took place a while ago, the MIT license was seen as more 
> straightforward and "obviously free" (nothing is free enough to some 
> die-hards) than the BSD license.
> 
> Or is:
> """
> Neither the name of the <ORGANIZATION> nor the names of its contributors 
> may be used to endorse or promote products derived from this software 
> without specific prior written permission.
> """
> 
> important to you in some specific manner?

Well, that clause isn't what's in my mind.  Rather,

a. scipy is BSD, so I'm happy just going with what they have.  I know MIT and 
BSD are very compatible, but in the future I think ipython and scipy are going 
to grow closer together, so the simpler the better.  I don't want to spend 
even 5 seconds of our time thinking we need to relicense anything to mix code. 
  If we all use the same license, we're good to go from day 1.

b. I already contacted the core authors about BSD, sent emails saying BSD, 
etc.  I really don't want to spend one more minute than absolutely necessary 
on this, so I'm not going to repeat that dance.

BSD stays.

Best,

f