|
Authored by: Anonymous on Tuesday, January 01 2013 @ 06:08 PM EST |
Is efficiency a determining factor in patentability?
No. But if the solution doesn't increase the efficiency, nobody is likely to try
to patent it.
Is doing something faster with nonpatentable components patentable?
No. But is it even possible to do? Very few hardware components are
nonpatentable.
[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Tuesday, January 01 2013 @ 10:10 PM EST |
First, consider www.mersenne.org. These guys use some very
sophisticated algorithms on zillions of copies of standard hardware to ask
questions in number theory like whether a given number with millions of
binary digits is prime, or, often more usefully, what are the factors of
numbers with a few hundred binary digits?
Like LZW compression, these algorithms were once new, are clearly
nonobvious, and sometimes useful, if you ave fast computers to run them
on. Should these be patentble?[ Reply to This | Parent | # ]
|
|
Authored by: Ian Al on Wednesday, January 02 2013 @ 05:17 AM EST |
The folk here are just going to moan, but here is my auto analogy, again. First,
a repeat of 35 USC § 101 - Inventions patentable:
Whoever invents
or discovers any new and useful process, machine, manufacture, or composition of
matter, or any new and useful improvement thereof, may obtain a patent therefor,
subject to the conditions and requirements of this title.
The
patent law does not say that one invention can be all of 'process, machine,
manufacture, or composition of matter' and the USPTO say that there can only be
one invention claimed in each patent application. So, one patented invention
cannot be a method and a system (aka machine) and a storage medium with software
installed in it (a typical mix of 'mirrored' claims in a so-called software
patent).
The precise definition set by the courts for a
'computer-implemented invention' is ' where software constitutes part of a best
mode of carrying out an invention'.
Fonar v. GE give the legal explanation
about the framework for inventions that include
computer-implementation:
As a general rule, where software
constitutes part of a best mode of carrying out an invention, description of
such a best mode is satisfied by a disclosure of the functions of the software.
This is because, normally, writing code for such software is within the skill of
the art, not requiring undue experimentation, once its functions have been
disclosed.
It is well established that what is within the skill of the
art need not be disclosed to satisfy the best mode requirement as long as that
mode is described. Stating the functions of the best mode software satisfies
that description test. We have so held previously and we so hold
today.
To explore the patentability, or not, of software I use the
example of a novel patented auto gearbox.
If the chosen control mode is
mechanical, then the patent is on the arrangement of gears, actuators, clutches
and the control mechanism. The invention is the arrangement of the components
and there may be some novel components that warrant a patent in their own
right.
You can replace the control mechanism with a computer (a
microcontroller) that electrically interfaces with the actuators. What is
patentable is the whole device and not the components. If the original
electro-mechanical control mechanism was patentable in its own right then it
would be the precise control function implementation that would be the patented
new machine.
Doing the precise control function with a computer should be
equally patentable. The software/software function would not be patentable, only
the computer/electrical interface/controlling function combination. Most
automatic gearbox control mechanisms will not be any more patentable than the
gears they manipulate so that even the special purpose computer assembly would
be the patent equivalent of a standard nut or bolt.
Although the
gearbox is controlled by a microcontroller, the magic is achieved by measuring
the gear speeds and electro-mechanically operating the levers. The actual
software in the microcontroller is irrelevant to the functions claimed for the
invention. All that is important is to state what functions related to engine
speed, output shaft speed, engine power and similar aspects of the design are
carried out by the software.
It is irrelevant what software language
is used to program the microcontroller and what software algorithms or functions
are used. If the same algorithms and functions are used with some other gearbox,
then the other gearbox only infringes if the inventive concept for gearbox
control in the patent is also used. The inventive concept is not in the software
functions carried out by the software, but is in the functions of the gearbox
controller.
In my example, the microcontroller is the best mode of
carrying out those control functions, but it could be replaced by mechanical
alternatives. The bit in Fonar that is almost universally ignored is that the
patented invention may be defined only by the functions if the software is only
a part of a best mode and, in that situation, it is only within the skill
of the art once its functions have been disclosed.
My auto gearbox
example (which I have been using for years to make this point) is the sort of
patent that Fonar encompasses. However, Fonar is clear that if software
constitutes all of the best mode for carrying out an invention, then
disclosing the function is not sufficient for a patent award. Also, if software
is all of the carrying out of the invention then by Benson, Flook,
Diehr, Bilski and Mayo the software process (or software machine) when
considered as a whole must be shown to have an inventive concept.
More seriously, it must be shown that the software, itself, is patentable
subject matter under §101’s implied list of excluded subject matter including
math, “laws of nature, physical phenomena, and abstract ideas.”.
In
other words, anything that is not a novel invention in the area of 'process,
machine, manufacture, or composition of matter, or any new and useful
improvement thereof' cannot be a patented invention. In Mayo, the Supreme Court
has expressed its fury that §101 was being skipped in the rush to
patenting.
... But in [the government's] view,
other statutory
provisions—those that insist that a
claimed process be novel, 35 U. S. C. §102,
that it not be
“obvious in light of prior art,” §103, and that it be
“full[y],
clear[ly], concise[ly], and exact[ly]” described, §112—can
perform
this screening function. In particular, it argues
that these claims likely fail
for lack of novelty under §102.
This approach, however, would make the “law of
nature”
exception to §101 patentability a dead letter.
The approach is
therefore not consistent with prior law. The
relevant cases rest their holdings
upon section 101, not
later sections. Bilski, Diehr, Flook, Benson. See also H.
R. Rep. (1952) (“A person may have ‘invent
ed’ a machine or a manufacture, which
may include any
thing under the sun that is made by man, but it is
not
necessarily patentable under section 101 unless the conditions of the title
are fulfilled” (emphasis added)).
We recognize that, in evaluating the
significance of
additional steps, the §101 patent-eligibility inquiry and,
say,
the §102 novelty inquiry might sometimes overlap.
But that need not always be
so. And to shift the patent
eligibility inquiry entirely to these later sections
risks
creating significantly greater legal uncertainty, while
assuming that
those sections can do work that they are
not equipped to do.
Using a formula or math algorithm in a computer is limiting that algorithm to
a particular technological environment. Computers only work because they can
execute software and the software has to be mathematically valid algorithms in
its entirety for it to work. All computer programs are math algorithms in each
and every part.
For a computer programmer, the variables become his
pets and the APIs become his toolbox. Only by an abstract visualisation of all
of the 'things' in software, can he write more software. Software starts to use
'objects', 'recognise' 'gestures' and 'play' music. It is not surprising that
this abstract world takes on the mantle of reality. Even software users believe
they are making 'gestures' control the computer. It's all clever stuff. It is
not patentable stuff, under the law.
The lack of understanding of how
computers work has lead to completely flawed judgements including some by the
Supreme Court. This article by PolR shows where and how the courts got it
wrong.
In Microsoft v. AT&T the Supreme Court opined that
AT&T's codec was manufactured by installing Windows Media Player in a
general purpose computer. In other words, WMP installed in a computer is an
infringing machine and not a 'new and useful process, manufacture, or
composition of matter, or any new and useful improvement thereof'.
The
installed Windows Media Player and computer was opined to be an infringing
machine because the implementation of the algorithm on an existing computer was
considered to be the invention of 'a new and useful' machine.
Fonar
tells us that the AT&T software on-a-computer is the whole of the 'best mode
of carrying out [the] invention' and that defining just the functions of the
invention (the math algorithm) results in an invalid patent. In the alternative,
the law tells us that the invention is just the method or process of using math
and the laws of nature and not a computer machine. Either way, it is not
patentable subject matter. As the Supreme Court said in Mayo v.
Prometheus:The Court has long held that this provision contains an
important implicit exception. “[L]aws of nature, natural phenomena, and abstract
ideas” are not patentable. Diamond v. Diehr, (1981); see also Bilski v.
Kappos,
(2010); Diamond v. Chakrabarty, (1980); Le Roy v. Tat-
ham, (1853); O’Reilly v.
Morse, (1854); cf. Neilson v. Harford, Webster’s
Patent Cases (1841) (English
case discussing
same).
and
[T]he prohibition
against patenting abstract ideas ‘cannot be circumvented by attempting to limit
the use of the formula to a particular technological environment’ or adding
‘insignificant postsolution activity’” (quoting Diehr),
The
AT&T patent asserted that the laws of nature are such that human hearing
does not require all of the data in a computer digital data file in order to
enjoy the content. The patent goes on to say that a math algorithm may be used
to reduce the unnecessary data. The codec patent was just the math of taking
the digital symbols representing audio and created using the CCITT international
standard and reducing the data so produced, but not required to store the audio
symbolised by the binary values. The only time this is a useful invention is in
a computer memory used for storing the data symbolising the original analogue
audio.
The patent says, use the math whenever using computers for audio
to get the best out of the laws of nature. It makes no claims about the software
or the processor or the computer sound card or the software language or the
operating system or the storage media.
The Supreme Court incorrectly
opined that when software (that used the laws of nature and included the
specific math algorithm cited in the patent) was installed in the memory of a
computer, that made a patent-protectable, useful, new
machine.
--- Regards
Ian Al
Software Patents: It's the disclosed functions in the patent, stupid! [ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Wednesday, January 02 2013 @ 11:07 PM EST |
First, consider www.mersenne.org. These guys use some very
sophisticated algorithms on zillions of copies of standard hardware to ask
questions in number theory like whether a given number with millions of
binary digits is prime, or, often more usefully, what are the factors of
numbers with a few hundred binary digits?
Like LZW compression, these algorithms were once new, are clearly
nonobvious, and sometimes useful, if you ave fast computers to run them
on. Should these be patentble?[ Reply to This | Parent | # ]
|
- No. - Authored by: jesse on Thursday, January 03 2013 @ 04:23 AM EST
|
|
|
|