Subject: Re: For Approval: Microsoft Permissive License
From: "Chris Travers" <chris.travers@gmail.com>
Date: Mon, 24 Sep 2007 13:57:30 -0700
Mon, 24 Sep 2007 13:57:30 -0700
On 9/24/07, Michael Tiemann <tiemann@opensource.org> wrote:
>
> On 9/22/07, Alexander Terekhov <alexander.terekhov@gmail.com> wrote:
> >
> > On 9/16/07, Chris Travers <chris.travers@gmail.com> wrote:
> > > Since the board is still considering this license, I wanted to provide
> > some
> > > information from my research that might help with this decision.
> > >
> > > One of the main issues people have on this list is the idea that other
> > > existing permissive licenses allow sublicensing.
> > >
> > > I am not a layer, but I believe that this is wrong as a matter of
> > intent
> > > licenses such as the BSD-licenses and the that it is also wrong as a
> > matter
> > > of law.  In fact, most permissive licenses do not allow for
> > sublicensing and
> > > force the licnese to follow the original copyrightable elements
> > (including
> > > but not limited to code).
>
>
> If the sublicensing is not required, then the safe harbour makes them feel
> permissive.
>


I am not sure I agree with this.  If, as I maintain, the BSD and even the
MIT licenses follow the source code elements* and the license cannot be
divorced from those elements, then I am not at all sure what makes the MS-PL
different from the BSD license.

* The MIT license provides permission to sublicense, but it also grants
permission to all downstream recipients to use according to its license, so
it superceded any sublicense, rendering it moot.  Thus even the MIT licensed
code is effectively governed by only one license and does *not* appear to be
compatible with GPL 3 section 7 because these additional permissions cannot
be arbitrarily removed by mere conveyance, but IANAL.

http://www.eweek.com/article2/0,1759,2185919,00.asp
> > (OSI Calls for Major Revisions to Microsoft Permissive License)
>
>
> I did not write this headline, and would not have characterized my
> comments as justifying such a headline.  I didn't call for anything, I gave
> a summary of what I read on license-discuss.
>

My only concern is that maximal incompatibility was no bar to approving the
GPL v3.   Hence it starts to look like we are approving submitters (FSF, MS,
etc) rather than licenses.

If, at the moment, the only measure we have accepted is the OSD, then IMO
this should have been approved at the same time.  Since the 2005 guidelines
have been abandoned (the GPLv3 clearly doesn't meet them), we also need a
replacement framework for dealing with license proliferation issues.

Best Wishes,
Chris Travers




On 9/24/07, Michael Tiemann <tiemann@opensource.org> wrote:
On 9/22/07, Alexander Terekhov <alexander.terekhov@gmail.com > wrote:
On 9/16/07, Chris Travers <chris.travers@gmail.com> wrote:
> Since the board is still considering this license, I wanted to provide some
> information from my research that might help with this decision.
>
> One of the main issues people have on this list is the idea that other
> existing permissive licenses allow sublicensing.
>
> I am not a layer, but I believe that this is wrong as a matter of intent
> licenses such as the BSD-licenses and the that it is also wrong as a matter
> of law.  In fact, most permissive licenses do not allow for sublicensing and
> force the licnese to follow the original copyrightable elements (including
> but not limited to code).

If the sublicensing is not required, then the safe harbour makes them feel permissive.


I am not sure I agree with this.  If, as I maintain, the BSD and even the MIT licenses follow the source code elements* and the license cannot be divorced from those elements, then I am not at all sure what makes the MS-PL different from the BSD license.

* The MIT license provides permission to sublicense, but it also grants permission to all downstream recipients to use according to its license, so it superceded any sublicense, rendering it moot.  Thus even the MIT licensed code is effectively governed by only one license and does *not* appear to be compatible with GPL 3 section 7 because these additional permissions cannot be arbitrarily removed by mere conveyance, but IANAL.

http://www.eweek.com/article2/0,1759,2185919,00.asp
(OSI Calls for Major Revisions to Microsoft Permissive License)

I did not write this headline, and would not have characterized my comments as justifying such a headline.  I didn't call for anything, I gave a summary of what I read on license-discuss.

My only concern is that maximal incompatibility was no bar to approving the GPL v3.   Hence it starts to look like we are approving submitters (FSF, MS, etc) rather than licenses.

If, at the moment, the only measure we have accepted is the OSD, then IMO this should have been approved at the same time.  Since the 2005 guidelines have been abandoned (the GPLv3 clearly doesn't meet them), we also need a replacement framework for dealing with license proliferation issues.

Best Wishes,
Chris Travers