Subject: RE: BSD-like licenses and the OSI approval process
From: "Lawrence Rosen" <>
Date: Fri, 12 Oct 2007 11:10:11 -0700

Alexander Terekhov keeps finding silly reasons to object:
> "When the collecting of the preexisting material that makes up the
> compilation is a purely mechanical task with no element of editorial
> selection, or when only a few minor deletions constitute an
> abridgment, copyright protection for the compilation or abridgment as
> a new version is not available. "

The creative aspects of my AFL 3.0-licensed collection of harvested
BSD-licensed works include my review of each of those BSD-licensed
components to ensure that:

 (1) the source code is indeed available to the public, as AFL 3.0
     requires even though the BSD licenses ignore this point; 

 (2) the BSD license text appears in the source code as the BSD 
     licenses require; 

 (3) the copyright notices are complete so as to accurately 
     identify provenance; 

 (4) the original BSD-like licenses are OSD compatible--in my 
     opinion and even without wasting time to obtain OSI's 
     blessing on those hundreds of old licenses; 

 (5) I'm comfortable offering my promise, as AFL 3.0 requires,
     that the software in my collection is "sublicensed to You 
     under the terms of this License with the permission of the
     contributor(s)...", knowing that the BSD-like licenses give
     me that permission; 

 (6) I have compared the code to other programs to ensure that 
     there aren't duplicate versions of the same programs that
     waste everyone's time; and 

 (7) I have undertaken other [proprietary!] techniques to make
     sure my compilation is effective, reliable and useful for
     the software commons. 

Then I would copyright my website text itself and my software that drives it
and that creates the database. I might or might not license this stuff under
OSL 3.0 and make a business of it.

If I was really sneaky, I'd patent this entire method for reducing all
software under BSD-like licenses into a single database of AFL 3.0-licensed
software for easy use. And I'd ensure that my patent claims were broad
enough to make my patent apply even if the collective license was something
other than AFL 3.0, for those of you who don't like that license. But damn,
here I've disclosed the method publicly. It is too late to apply for a
patent, at least outside the US. So anyone who thinks there is value in
reducing all those old BSD-like licenses to a single modern academic
license, feel free to do it. :-) [Be warned: In the US, I believe I still
have one year to file this business method patent. But by this disclosure, I
can at least ensure that YOU won't try to patent it!]

Copyright works!


Copyright (C) 2007 Lawrence Rosen.
This email (with the exception of the quotation from Alexander Terekhov) is
licensed under the Open Software License (OSL 3.0).