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

[Nicolás A . Ortega]

On Mon, Apr 17, 2017 at 07:33:23PM -0400, Michael Pagan wrote:
> On 17 Apr 2017 at 14:22, Nicolás A. Ortega wrote:
> >> On Sun, Apr 16, 2017 at 11:32:58PM -0400, Michael Pagan wrote:
> >> <SNIP>
> >> This is the direct strategy employed by the GNU Project.  I believe RMS
> >> got it right when he decided that the license that should be chosen for
> >> a library should be based on whether it provides an advantage or not for
> >> the free software community.  If the library does not provide an
> >> advantage (i.e. the library shares the same features as non-free
> >> libraries), then a "weak" license should be used; if the library *does*
> >> provide an advantage (i.e. pioneering a new significant feature), then a
> >> copyleft license should be used.
> >> 
> >> Learn more about this strategy here:
> >> http://www.gnu.org/licenses/why-not-lgpl.html
> >> 
> >
> > The issue with this kind of method is that many people are put off by
> > the idea of using a GPLed library (mostly because that brings up a ton
> > of other legal concerns about the other libraries that you may be
> > using), so it makes things extremely messy. LGPL keeps the library code
> > Free, but does not ensure (legally) that the library is only used by
> > Free Software.
> 
> I think I see where you're going here.  Could it be that a copyleft
> library can't be used with non-copyleft libraries in parallel?  Do you
> mean that if someone "copylefts" a library that it will no longer work
> with a non-copyleft library in a legal sense?  I am excluding the LGPL
> From these questions when I mention /copyleft library/.
> 
> The key question here is: Does it really matter?  The copyleft library
> can still be used, and if it does not work with other libraries, then
> those libraries can either be: Substituted with copyleft ones, if they
> exist; refactored into a single copyleft library; or maybe even forked
> (as a last resort).
> 

At which point your forcing someone to perhaps refractor their entire
project for the sake of one library, so unless your library is
irreplaceable they will end up simply not using your library. And even
if it is irreplaceable at the moment, it won't be as soon as someone
makes a permissively licensed version of it.

> The above may sound like a duplication of effort, but if any of those
> non-copyleft libraries ever become non-free (not saying that they will),
> then the TRUE duplication of effort will begin (i.e. reverse-engineering
> the proprietary features, and adding them back into our free library).
> 
> I'd rather have a guarantee of freedom (i.e. copyleft), then simply
> freedom in the moment (i.e. free, but non-copyleft), no matter what the
> cost; even if legal hassles and certain sacrifices (no non-copyleft
> libraries, which may be a stretch for many) must be made.
> 
> Perhaps you are seeing something else of significance that I do not?
> Maybe you have envisioned a scenario where using a copyleft library in
> specific circumstances makes certain types of computing more difficult
> due to some computational sacrifices.
> 

I think you are misunderstanding what I am looking for. I am completely
in favor of copyleft, however I don't think it's up to me to decide what
license someone who uses my library should use, yet I do not wish to
help anyone making non-free software. Before I simply thought I'd use
the LGPL, since it is copyleft for derivatives, but not any software
that links to it.

> > Nope, goes a little further than that. To my understanding (however it
> > seems that I may have been wrong) Sleepycat is copyleft upon derivatives
> > and forces source disclosure on software that uses the Sleepycat
> > licensed software.
> 
> I have read the license.  _The Sleepycat license is non-copyleft_.  It
> is most definitely a lax, permissive "weak" free software license.
> 
> Sleepycat behaves just like the non-copyleft Modified BSD License; thus,
> it is also non-copyleft.  For further clarification, let's dissect the
> license-- shall we?
> 
> Here it is in part.  I eliminated the copyright notice and duplicate
> clauses (clause #3 is removed, since it deals with attribution, which is
> not our main focus right now).  The first 2-clauses deal with the
> distribution of a hypothetical program:
> 
>     * Redistribution and use in source and binary forms, with or without
>     * modification, are permitted provided that the following conditions
>     * are met:
>     * 1. Redistributions of source code must retain the above copyright
>     *    notice, this list of conditions and the following disclaimer.
>     * 2. Redistributions in binary form must reproduce the above copyright
>     *    notice, this list of conditions and the following disclaimer in the
>     *    documentation and/or other materials provided with the distribution.
> 
> Because the conditions (clause #1 and clause #2) only deal with
> "redistributions" (i.e. exact copies), "derivative works" or "modified
> versions" are not affected by the license.  Redistribute means to
> distribute again or anew.  The following represents a hypothetical
> revision that would need to take place in order for the Sleepycat
> license to be copyleft (the changes are in ALL CAPS):
> 
>     * 1. Redistributions of source code, AS WELL AS MODIFIED VERSIONS,
>     *    must retain the above copyright notice, this list of conditions
>     *    and the following disclaimer.
>     * 2. Redistributions in binary form, AS WELL DISTRIBUTIONS OF
>     *    BINARIES PRODUCED THROUGH THE COMPILATION OF A MODIFIED VERSION
>     *    OF THE SOURCE CODE, must reproduce the above copyright notice,
>     *    this list of conditions and the following disclaimer in the
>     *    documentation and/or other materials provided with the
>     *    distribution.
> 
> In parenthesis, I'm making it clear the obscure way that the Sleepycat
> license and the Modified BSD license provides the 4 essential freedoms:
> 
>     * Redistribution (FREEDOM #2) and use in source (FREEDOM #1) and
>     * binary forms (FREEDOM #0), with or without modification (FREEDOM
>     * #3), are permitted provided that the following conditions are met:
> 
> 
> Your thoughts?

Yes, it seems as though I had misinterpreted the license. However the
concept that I thought Sleepycat would provide can more or less be
expressed how I did so in another branch of this thread[0]. In any case,
yes, you are correct from what I can see and make very good points.

[0] https://lists.gnu.org/archive/html/libreplanet-discuss/2017-04/msg00019.html

-- 
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