[ros-users] manifest.xml license tag

Bill Morris morris at ee.ccny.cuny.edu
Thu Jun 16 22:54:34 UTC 2011


On Thu, 2011-06-16 at 15:23 -0700, Ken Conley wrote: 
> On Thu, Jun 16, 2011 at 3:06 PM, Bill Morris <morris at ee.ccny.cuny.edu> wrote:
> > On Thu, 2011-06-16 at 14:43 -0700, Ken Conley wrote:
> >> It's a simple addition, but I'm wondering if it is redundant with the
> >> URL.  i.e. while there is a general URL for the LGPL license, there is
> >> a specific URL for the 3.0 license:
> >>
> >
> > That is a reasonable solution, but not as machine parseable. I can think
> > of use cases where someone doesn't want to install GPLed code, but I'm
> > not sure there is a use case for the version info.
> 
> Basically I'm wondering if
> 
> <license url="http://www.gnu.org/licenses/lgpl-3.0.txt"
> version="3.0">LGPL</<license>
> 
> is preferable to
> 
> <license url="http://www.gnu.org/licenses/lgpl-3.0.txt">LGPLv3</<license>
> 
> The separate version attribute favors the family of license, names the
> license as a physical entity.  I'm mildly leaning towards the latter
> because license versions are not really equivalent to software
> versions.  GPLv3 really is a distinct and different license from
> GPLv2, it's not an upgrade per se.

The advantage with the idea of grouping licenses is that decreases the
number of permutations. However, I'm not sure what use there is for a
machine parseable license tag besides code audits.

> Also, I would include "please, please don't use Creative Commons.
> Even Creative Commons says don't do it":
> http://wiki.creativecommons.org/Frequently_Asked_Questions#Can_I_use_a_Creative_Commons_license_for_software.3F

Another problematic one is undefined licenses. Anything is better than
undefined because I can't agree or disagree to an undefined license.

> >> ???
> >> No Clue
> >> Unknown1
> >> WhoCares
> >> trivial-features
> >> unknown






More information about the ros-users mailing list