Subject: Re: termless copyright and patents
From: Thomas Lord <>
Date: Sat, 07 Oct 2006 10:37:30 -0700 wrote:

>  > You seem to want to believe that if the program Alice conveys
>  > is captured by scope assertions in "claim #2" of her patent,
>  > that therefore she has forgiven all possible infringement under
>  > claim #2 in the conveyed work or any derivitive.
> That's right.

A patent contains a set of enumerated claims and, collectively,
these carve out the overall scope of the patent.   In some
contexts -- especially in covenants -- "essential patent claims"
refers to an arbitrary subset of that scope.

To illustrate, here is one list of claims, one
independent and two dependent:

     1. a frob

     2. a frob, as in (1), which is red and has three legs

     3. a frob, as in (1), which is blue and has four legs

One subset of the scope which can not be constructed
by picking and choosing from the enumerated claims

      implied-1a.   a frob as in (1) which is blue and has three legs

As the term is commonly used by standards organizations
and in covenants, "essential patent claims" refers to
claims like "implied-1a".     If Sun submits an RFC
to the "three-legged blue frobs" standardization committee
they are making a covenant not to assert their essential
patent claims for three-legged blue frobs.   They are not
promising not to assert against "legless blue frobs".  They
are not promising not to assert against "orange, 7-legged
frobs".    The essential patent claims which Sun is covenanting
are not described by (1) or by any combination of (1), (2),
or (3):  the essential patent claims are a *subset* of (1).

And that makes perfect sense.   The purpose of the
*enumerated* claims is to stake out the boarders of
the claimed scope.    It is *not* the purpose of the
enumerated claims to limit the patentee into how that
territory of claims can be divided up.

If you claim a square acre, that is the same thing
as having 4 separate claims to each of the four quads
or 16 claims to the subquads of those.   The description
of the boarders on the deed are one thing but the "essential
parcels" to which you can give or deny permission for some
use are different.

Suppose that Sun has many patents that are essential to
"NFS 6.0".    They release a GPLv3d2 program which is
an NFS 6.0 server.    The patent doesn't mention NFS 6.0
at all (and was issued before NFS 5.0 was even finished).
As far as a standards body is concerned, the essential patent
claims of Sun are limited in scope to what is essential to
practice the proposed standard -- they don't imply, for example,
the right to practice the patents in another network file-system.
Where is the language that makes GPLv3d2 any different?

>  > A patentee can not induce infringement and then turn around
>  > and sue for the infringement they induced.  This is why, for
>  > example, you see some classes of product marked with patent
>  > numbers or marked "patent pending".
> Highly unlikely, although I don't have time to read 35 USC right now.
> (Modulo my temporary ignorance) the difference is not whether you can
> sue for infringement (cf the LZW patent), but whether the infringement
> will be classed as inadvertent (temporary injunction to stop and/or
> payment of back license fees) or deliberate (treble damages).

Between estoppal, the definitions of inducement and liabilities of inducers,
and the provisions of GPLv2, I think a "suggested patent claims"
covenant is strongly implied by GPLv2.