Crafting a special kind of license for a very special standard.

Matthew Flaschen matthew.flaschen at gatech.edu
Sat Apr 14 18:15:56 UTC 2007


Grant Robertson wrote:
> I really want the assistance from as many in the open-source community
> as possible so that I can guarantee that the license will both protect
> the standard and promote the development of a new family of
> open-source software. The home page can be found at http://demml.org
> and the page about licensing issues can be found at
> http://demml.org/standard/license/.

IANAL, but I think it is meaningless to specify a license for a standard
itself.  The standard document is copyrighted, and that can have a
license.  Also, techniques required by the standard can be patented, and
those too may have licenses.  You should understand that the document
copyright can not be used to stop undesired third-party implementations.
 Someone with a copyright license for the document can describe it in
their own words, then distribute the resulting document freely; facts
can not be copyrighted.  Only patents can control implementation.  Since
you do not seem to intend to patent any required techniques, I fail to
see how you can control implementations.  Publishing prior art can stop
others from patenting your ideas, but it does absolutely nothing to
prevent unauthorized implementations.

Your licensing scheme is also clearly not open in the way Rosen
described.  Charging fees for commercial use violates his second
principle, "Open standards should be available to everyone on
royalty-free terms."  There are other problems with your terms (like
requiring beta releases of implementing software to be hidden), but that
alone makes your standard impossible to implement with open source
(which /can/ be commercial).  You also say you intend to submit your
standard to W3C, which I believe requires standards can be implemented
royalty-free
(http://www.w3.org/Consortium/Patent-Policy-20040205/#sec-Licensing).
Then again, you don't seem to have any patents, so it's a moot point...

Matthew Flaschen



More information about the License-discuss mailing list