COMPUTER IMPLEMENTED INVENTION: they are inventions that are based on a logic operation (program) of a microprocessor or equivalent logic unit.
PROGRAM: typically is a set of instructions in the memory of a microprocessor that, being performed, modify operating internal or external parameters to the microprocessor. The definition must also include "embedded" logic programs in ROM, eg. as ASIC or FPGA, or discrete combinatorial networks (eg. a circuit on a PCB).
Contrary to many misinformed rumors, SOFTWARE patents exist outside the U.S. as well, a country that historically has fostered and promoted their spread. Not surprisingly, the U.S. patent law allows to protect business method by patents, which is prohibited in Europe.
In other words, even in Europe SOFTWARE is patentable, and since a long time!
To avoid falling into confusion, it is immediately necessary to clarify WHAT is meant by patentable SOFTWARE. The Industrial Property Law expressly prohibits computer programs, but as such. The words "as such" must be so interpretated: only the algorithms in their abstract essence are excluded from patenting. That is, phases of calculation or logical expressions may not be protected when considered in the mind of the programmer, on a sheet of paper or on a blackboard. All these manifestations of intellectual, sequential or logical-mathematical reasoning can not be monopolized. But their direct translations into practice can.
Let's take an example. If someone invents a new calculation algorithm to create a digital filter, eg. by combining ingeniously a series of additions and multiplications, so that the computational cost decreases, unfortunately he can not patent that algorithm itself. Otherwise no book could mention it, or no university could teach it, and scientific progress would stop. However a practical application of the algorithm, eg. a DSP loaded with instructions that perform the algorithm, can (and should) be patented!
If the practical embodiment of an abstract concept is patentable, a company or an entrepreneur can be well satisfied. We want to say that despite the limitations "on paper" of the patentable scope for the software, what remains protectable covers in practice all commercial or economically attractive alternatives. If a company invents a new program for the control of a computer-controlled machine will not be much angry upon not being able to protect the algorithm itself, but will be satisfied instead by protecting: every machine that uses the program, the program controlling the machine and a support with the program stored therein.
In the previous sentence we have just written "program", it is not a mistake. One wonders if there is an inconsistency with what was said earlier. Well, to dispel any doubts, we add that a program is protectable in itself as long as the execution gives a technical and tangible result. It is clear, then, how the spectrum of industrial applications is considerably more spacious than it seemed at the beginning.
Returning to the example of the digital filter, that algorithm is not protectable in itself if its computational result is eg. another filter (and therefore an abstract entity) or eg. a calculation of coefficients by a faster method. But if the algorithm is able to give an output of samples (and thus an electrical signal physically detectable) with some peculiarities, one can protect not only a DSP which contains the program, but also the program loadable on this DSP. Obviously a new filter would be protectable as a numerical method performed by a DSP on REAL samples.
The watershed that divides the patentable algorithms from those excluded is in essence the ability, or the effective application, of the algorithm to give a result that is applied or applicable to the physical or industrial world.
To make good SOFTWARE patents, more than for other branches of technology, a good technical knowledge (electronics, computer science, systems engineering) is necessary.
CB firm contains the whole experience of engineering and APPLIED research acquired not only by a degree in electronic engineering, but also from years of R&D, research and doctoral specialization and continuous curiosity.
This is combined with strong technical knowledge of patent law, in every branch and shade for the various patent laws in the world. Obviously it would be a mistake to write a patent in the same way and with the same techniques knowing that its text will be evaluated by national examiners with different rules. Therefore, not only from the customer the expected territorial horizon which the patent has to cover is evaluated, but the patent will be written trying to meet and satisfy all national requirements. Only in this way a defined, comprehensive and uniform scope of protection is ensured (without too many territorial changes).