|
|
|
| Authored by: Anonymous on Friday, January 04 2013 @ 02:48 PM EST |
I disagree that there is no room for software patents. I agree that the current
system is abused, but I think there's a subset of software patents that
actually
do make sense. Processes that pass muster for method patents
(including but
not exclusive to the machine-or-transformation test) shouldn't
become
unpatentable just because they happen to be implemented on a computer.
The constitutional foundation for patents exists explicitly for the purpose of
protecting inventors from having their work appropriated before they can
market it, and prohibiting software patents outright opens up the window for
an inventor's product to be cloned by one of the software giants and brought
to market for far less cost, thus disincentivizing independent investment in
novel development and thereby failing to "promote the Progress of Science
and
useful Arts" as the Constitution requires.[ Reply to This | Parent | # ]
|
| |
| Authored by: reiisi on Friday, January 04 2013 @ 07:28 PM EST |
There are several issues you bring up which I want to emphasize.
First, not very many software inventions meeting the uniqueness requirement are
simple enough to be understood and reconstructed without the documentation, real
source code, specifications for the compile and run-time environment, and
minutes of the design meetings.
Second, very little software meeting the uniqueness requirement is actually
understandable by a single person.
We (as a current culture) make a serious mistake in arguing that complexity
should equate to patentability. It is precisely the reverse when things get this
complex.
And almost anything that can be understood by a single person has already been
part of the state of the art since practically before the industry began. (In
other words, the mathematical underpinnings, the extant business and industry
practices and methods, etc. And the argument/claim of "implemented on a
computer" has to be substantiated by specifying which computer, with which
run-time, compiler, interpreter, etc., before it can be called a newly invented
machine. And the specific machine claims have to be considered limitations to
the patent, or it's back to being a patent on ideas.)
Which is part of the "software is literature, and copyright is the
appropriate tool" argument.
[ Reply to This | Parent | # ]
|
| |
| Authored by: Ian Al on Saturday, January 05 2013 @ 08:03 AM EST |
Is 'a process or method implemented on-a-computer' always a 'software patent'?
When the patent claims are purely functional (in abstract terms, not software
function terms) is the patent adjudicator supposed to guess based on the
disclosures and the names put on the diagrams? (Note that only the claims are
the legally patented invention).
The patent lawyers art is such that any rejection by software coders can be
challenged by asserting that the claims on which they are basing their rejection
are not stated to be implemented by software. in fact, the lawyers would quickly
undermine the process by all sorts of legal challenges including that the
functional claims never mention software implementation.
I agree with you that the test is legitimate on the basis of Fonar, but the vast
body of dubious 'software' patents are ignoring the legal details of Fonar. The
patent lawyers are just too good at avoiding the law.
---
Regards
Ian Al
Software Patents: It's the disclosed functions in the patent, stupid![ Reply to This | Parent | # ]
|
|
|
|
|
