|
Authored by: PolR on Thursday, January 31 2013 @ 08:52 PM EST |
It is my understanding that non-working pseudo code would lead to an invalid
patent. The invention must be "useful". This is a requirement of patent law. An
invention that can't be actually built is not useful.
Also section 112
paragraph (a) of patent law states: (my bold)
The specification
shall contain a written description of the invention, and of the manner and
process of making and using it, in such full, clear, concise, and exact
terms as to enable any person skilled in the art to which it pertains, or with
which it is most nearly connected, to make and use the same, and shall set forth
the best mode contemplated by the inventor or joint inventor of carrying out the
invention.
So unimplementable pseudo-code violates section 112(a)
and the patent is invalid.
IANAL by this is how I understand these things
would work.
[ Reply to This | Parent | # ]
|
|
Authored by: myNym on Thursday, January 31 2013 @ 11:58 PM EST |
I'd like to see working code, complete with build
instructions as to which compiler and libraries to use, and
on which platform said code would execute.
Code doesn't run as claimed, patent is invalid.
(There are no valid software patents in my opinion anyway.
I'm referring to Diamond type patents where the software is a
non-patentable component of a presumably patentable process.)[ Reply to This | Parent | # ]
|
|
Authored by: nick_battle on Friday, February 01 2013 @ 03:16 AM EST |
There is a mid-point between informal pseudocode and a working program, namely a
formal specification (http://en.wikipedia.org/wiki/Formal_specification).
Formal specification languages eliminate the ambiguity of pseudocode, and allow
the pure algorithm to be analysed independently of how it may be implemented in
a programming language. This seems to be precisely what we are looking for
here.
Unfortunately, writing and reading such specifications does require considerable
skill. But then who ever said specifying a patent should be easy? :)
[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Friday, February 01 2013 @ 12:53 PM EST |
The huge problem with functional code is copyright. Actual
code in a patent
would probably be covered by copyrighted
even after the patent has expired. So
that would defeat the
purpose of patents. Releasing the code under some license
might work but that license cannot conflict with the patent
laws.
Also the
laws would have to change to address porting
between languages. That is, if you
patent was just in the C
language then the code clauses would not apply to
another
language like C++. [ Reply to This | Parent | # ]
|
|
|
|
|