Motivation for Sleepycat + MIT hybrid

[Matthew Flaschen]

John Cowan wrote:
> Matthew Flaschen scripsit:
> 
>> [The LGPL] can be linked into any program (even proprietary
>> ones), as long as the program's license allows private modification
>> (however, source need not be provided) and reverse engineering for
>> debugging those modifications.  Please read LGPL section 6
>> (http://www.gnu.org/licenses/lgpl.html).
> 
> IMAO, the LGPL is the most incomprehensible widely used FLOSS license
> (the *most* incomprehensible FLOSS license being the Reciprocal Public
> License).

Perhaps, but I believe it serves a necessary purpose.  The MPL's
file-by-file test is certainly simple, but essentially makes the
copyleft worthless (#include anyone?).  The LGPL attempts to preserve it
where possible.

  And while the matter is not subject to proof, I believe
> that most creators of proprietary software either avoid LGPLed code
> because the license is so complicated, or else simply ignore the
> requirements for private modification and reverse engineering.

It's really not that complicated.  People probably do violate the
private modification and reverse engineering requirements, but they only
take one sentence in the proprietary license to do right.  The hard part
is figuring out the difference between modifying the library and using
it in a new application; there's no easy solution here.

> IOW, the folk theorem about the LGPL is "Just like the GPL, except
> you can use it in proprietary stuff too."

That is true to an unfortunate extent.  Education is the answer.

Matthew Flaschen