> Just reading New Scientist i read in the feedback column this
> "Patents guru Greg Aharonian runs the Internet Patent News Service [No
> Address Given] and each year tots up the statistics on software-related
> patents...."
>
> I therefore assume this is the curent status and you can patent software.
> The issue is 4 Jan 1997 No2063.
>
> Tim
>
With all due respect, it is appropriate in a case such as this to "go
to the horse's mouth" to get a proper definition.  I'm afraid that Mr.
Aharonian's opinion, or, worse yet, inferences drawn from his choice of
words may be misleading.

I have gone to the U.S. Patent and Trademark Office at www.uspto.gov
to find the current definition of what is patentable under U.S. law.
The following is taken from that site under the heading "What can be
patented":

   The patent law specifies the general field of subject
   matter that can be patented and the conditions under
   which a patent may be obtained.

   In the language of the statute, any person who "invents
   or discovers any new and useful process, machine,
   manufacture, or composition of matter, or any new and
   useful improvements thereof, may obtain a patent,"
   subject to the conditions and requirements of the law.
   By the word "process" is meant a process or method, and
   new processes, primarily industrial or technical processes,
   may be patented.  The term "machine" used in the statute
   needs no explanation.  The term "manufacture" refers to
   articles which are made, and includes all manufactured
   articles.  The term "composition of matter" relates to
   chemical compositions and may include mixtures of
   ingredients as well as new chemical compounds.  These
   classes of subject matter taken together include
   practically everything which is made by man and the
   processes for making the products.

Obviously, the category above that comes closest to software is
"process."  But, note that the given definition of a "process" refers
to a process or method for doing [and here one must presume] something
else.  As I indicated earlier, a couple of years ago this was the
interpretation of the courts.  I doubt very much (though I readily
admit I don't know with absolute certainty) that this interpretation
of a "process" as meaning a method for accomplishing some other end
has changed significantly in the interim.  It is the sort of issue
that, had it so changed, would have been a topic of much public
debate.

Perhaps the best way to gauge the current interpretation is to look at
a recently issued U.S. patent that is alleged to be a "software"
patent.  I think you will find that the software part of the patent is
described always in association with either a process it controls or a
piece of hardware which it likewise controls.  I doubt that you will
find that the software is patented all by itself with no connection or
reference to any controlled process or hardware.

If anyone is aware of a recent "software" patent that makes no
reference to a controlled process or hardware, please bring it to my
attention.  I have at least one powerful mathematical algorithm I have
discovered that I would like to patent that does not lend itself to
the control of a process or machine.  I would very much like to see
how such a patent is written.

Thanks,
--- Warren F. Davis
================================================
Davis Associates, Inc.
43 Holden Road
West Newton, MA 02165  U.S.A.

Tel: 617-244-1450        FAX: 617-964-4917
Visit our web site at:  http://www.davis-inc.com
================================================