Authored by: Anonymous on Thursday, January 31 2013 @ 09:44 PM EST |
Mayhap add the requirement that a copy of WORKING, COMPILABLE
SOURCE code implementing the software parts of a patent
accompany the application. since it's digits, it doesn't
take physical storage space (the original reason for
abandoning the requirement for a working model of a patent).
That reference implementation must be published as part of
the patent.[ Reply to This | Parent | # ]
|
- Pseudo-code - Authored by: Anonymous on Thursday, January 31 2013 @ 11:09 PM EST
- Pseudo-code - Authored by: Anonymous on Friday, February 01 2013 @ 04:50 AM EST
- Pseudo-code - Authored by: Anonymous on Friday, February 01 2013 @ 04:51 AM EST
- Pseudo-code - Authored by: om1er on Friday, February 01 2013 @ 07:19 AM EST
|
Authored by: Anonymous on Thursday, January 31 2013 @ 09:54 PM EST |
Thank you for the reference to 112 (3). The point I was trying to make was
that pseudo-code may not accurately disclose the invention. It simply discloses
what the troll wants to claim the invention does. I maintain that
all pseudo-code is non-working. It is simply a means of communicating
what someone wants the actual code to do.
Being a level of
abstraction away from the invention it is subject to abuse by not actually
disclosing what the real code does and it may still not disclose anything
useful.
I have reviewed pseudo-code written by analysts from the
business side as well as pseudo-code written by analysts from the development
side. Using the pseudo-code written by a developer usually results in real code
closely matching the pseudo-code. Not so much for pseudo-code written by the
business side.
In the context of patten law I don't see pseudo-code
as being significantly better than the legal descriptions we have today. Trolls
will start using vague descriptions in business type pseudo-code to obtain
pattens that are no better than what we have today.
IANAL
cc
[ Reply to This | Parent | # ]
|
|
Authored by: bugstomper on Thursday, January 31 2013 @ 10:32 PM EST |
Look at what is accepted in existing software patents and how much time and
money is wasted in the courts because there is no way of knowing what the claims
actually describe. As far as I can tell 112(a) is essentially ignored when it
comes to software patent fights. It's tailors fighting over the design patents
for the Emperor's clothes - None of them will want to try to win the battle by
pointing out that there is nothing there.
If 112(a) is so eviscerated for software patents now and the solution is to
require source code, then it needs to be working source code. Proper pseudo code
is the same as working code but you don't know that it is proper until you
actually try to translate it to a real language and run it.
[ Reply to This | Parent | # ]
|
|
Authored by: BitOBear on Thursday, January 31 2013 @ 11:53 PM EST |
All pseudocode is non-working. That's what the "pseudo" means.
How to be a Millionaire and never pay taxes() :
Get Million Dollars;
Don't pay taxes;
End;
Pseudocode is intended to disclose structure, not implementation. It is the very
thing we allege to dislike in Section A. It is (usully) a description of
function without algorithm. Some people (mistakenly) nearly-code (my own coinage
of phrase) when pseudocoding by more or less dropping the punctuation, but this
is not the norm nor the purpose.
Pseudocoding has largely evolved into (e.g. been replaced by) data flow diagrams
and structural diagrams for the very reason that the archetypal pseudo code is
not reasonably informative in most cases.
The internet meeme with step N being "???" and step N+1 being
"profit!" is basically pseudocode with the "magic happens
here" verb that most pseduocode is plagued with.
For example "get data; sort data; find best fit; done" had lots of
magic in the third step.
In short everything _but_ actual execution-ready code is capable of hiding
unbounded vagueness. This includes all pseudocode and pseudocode-alikes such as
UML.
The patent office used to require a working example of the invention. I think
they stopped because they had no place to put it all. But in software terms,
there is no substitute for a working invention because any "natural
language" (e.g. English etc) verb is potentially NP incomplete.[ Reply to This | Parent | # ]
|
|