#1 is why the current U.S. law extending patents to "business methods"
is so questionable, as those holding those patents seem wont to lay
claim to the algorithms underlying those methods, not to a specific
physical implementation for that algorithm. An algorithm is more like
a creative expression than like a physical invention, and as such should
be treated differently than a physical invention, The only requirement
for creation of an algorithm is the equivalent of pencil and paper and a
mind, which both increases the possibility of concurrent, independent
invention and complicates the depth of search required to rule out prior
art.
JC Ewing
On 01/12/2013 12:30 PM, Charles Mills wrote:
Simply, No. For at least three reasons.
1. Copyright and inventions are different domains. Copyright protects a
particular creative expression -- think of sonnets. Inventions are ways of
doing things. Think cotton gin. They are different domains.
2. Only the inventor can patent something. One cannot read about something
nifty and run out and patent it; one has to BE the inventor.
3. "Prior art" precludes obtaining patents (or, if push comes to shove, are
a defense against the enforcement of a patent that turns out after issuance
to be invalid because of prior art). Even if I wake up this morning and
innocently "invent" something, if it turns out that you wrote a paper about
it a couple of years ago then I am precluded from obtaining (or the courts
would invalidate the patent and preclude me from enforcing) a patent on my
invention (even if I never read your paper).
I think this is pretty much what John is saying. I don't see how what John
wrote could be interpreted as saying "inventing a software piece,
copyrighting it ..., a third party can patent the invention ..." What John
says is the opposite: "the putative 'invention', being already in the public
domain, is non-novel and thus not patentable."
BTW, copyright does not imply "public domain."
1. You mean "publically accessible." "Copyrighted and "public domain" are
essentially antonyms.
2. See my other note on copyright. No registration (public disclosure) is
necessary at all.
3. You can register a copyright with key elements redacted for trade secrecy
reasons.
Charles
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On
Behalf Of Graham Hobbs
Sent: Friday, January 11, 2013 4:42 PM
To: [email protected]
Subject: Re: OT: IBM #1 in number of patents for 2012. It's 20th year in a
row to do so.
John,
Could one conclude then, that inventing a software piece, copyrighting it
(thus on a Govt database thus a public domain), a third party can patent the
invention and you could end up paying the third party to use your own
software?
I read your comments closely and the above cannot happen!+ ..?
Graham
----- Original Message -----
From: "John Gilmore" <[email protected]>
Newsgroups: bit.listserv.ibm-main
To: <[email protected]>
Sent: Friday, January 11, 2013 12:49 PM
Subject: Re: OT: IBM #1 in number of patents for 2012. It's 20th year in a
row to do so.
Graham,
"Prior art" is a lawyer's term, not mine. If, say, I published a
paper in 1977 describing a scheme for extracting the square root of a
non-negative integer by successive subtraction, using the fact that
n^2 is the sum of the first n odd integers, i.e.,
2^2 = 4 = 1 + 3
3^2 = 9 = 1 + 3 + 5
4^2 = 16 = 1+ 3 + 5 + 7
5^2 = 25 = 1 + 3 + 5 + 7 + 9
. . .
a later attempt to patent such a scheme would [almopst certainly]
fail. This relation is due to Fr Marin Mersenne (1588-1648), but
while no mathematical relation is itself patentable a device based
upon it, which might otherwise be patentable, would not be if my paper
describing such a device were known to the relevant Patent Office.
Thus, while my paper might well be copyrighted (most published papers
are), that is not the point; what is crucial is that the putative '
invention', being already in the public domain, is non-novel and thus
not patentable.
--
Joel C. Ewing, Bentonville, AR [email protected]
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
