Re: [libreplanet-discuss] Potential of the Sleepycat License

[Nicolás A . Ortega]

On Sat, Apr 15, 2017 at 04:48:52PM -0700, Ian Kelling wrote:
> 
> Nicolás A. Ortega <deathsbreed@themusicinnoise.net> writes:
> 
> > I've tried having this discussion on #fsf and #gnu, and I think that
> > this license has the potential to be a great software license,
> > especially for libraries.
> >
> > To my understanding the Sleepycat License[0] is a copyleft license in
> > which all derivatives of the work must be licensed likewise (under the
> > Sleepycat license) and works that use a project under this license must
> > disclose source code.
> >
> > There are, however a couple problems with this license, the first one
> > (as you most likely have noticed while reading the above) is that
> > disclosure of source code does not mean free software, and secondly is
> > the issue that the license uses very specific terminology referring to
> > the BerkleyDB (the software that uses this license) and refers mostly to
> > DB software. Given, disclosure of source code is better (imo) than the
> > LGPL since it forces the disclosure of the sources (while LGPL only does
> > so in the case of static linking if there is no exception), and still
> > gives more freedom for the programmer to choose a license unlike one of
> > the GPL licenses (despite how much I love them).
> >
> > However, if we can find people with the knowledge to write/modify
> > licenses ('cause I for sure will not be able to do that) then I think
> > that this license could be modified to fix those two problems (for
> > example, instead of requiring that code be disclosed, all 4 freedoms
> > could be required).
> >
> > I am not an expert in licensing, which is why I brought this up here.
> > Hopefully someone here has the ability, time, and will to do this (if it
> > is possible). (^_^)
> >
> > [0] https://en.wikipedia.org/wiki/Sleepycat_license
> 
> A modification can't "improve" the license, because any modification
> simply makes a new license.  This can be mitigated if the old license
> has an automatic upgrade provision, but sleepycat doesn't. A new license
> comes with significant downsides, see
> https://en.wikipedia.org/wiki/License_proliferation. Without some
> specific compelling reason an existing license wouldn't work, it's
> better not to write a new license. I'm glad you are thinking about
> giving people freedom, unfortunately, I don't think this is a good way
> to do it.

I looked at the link you were mentioning, I had never heard of License
proliferation before, so now I understand the issues with doing
something like this.

In which case, I am wondering if it would be possible in such a case to
use a combination of FLOSS licenses to create something similar? For
example, perhaps one could license a library under the (A)GPL and then
offer LGPL (with static linkage exception) dual-license to Free Software
projects.

-- 
Nicolás Ortega Froysa (Deathsbreed)
https://themusicinnoise.net/
http://uk7ewohr7xpjuaca.onion/
Public PGP Key:
https://themusicinnoise.net/deathsbreed@themusicinnoise.net_pub.asc
http://uk7ewohr7xpjuaca.onion/deathsbreed@themusicinnoise.net_pub.asc

signature.asc
Description: PGP signature