OSNews

I've been writing code for over 25 years. Code is not mathematics. Far from it. Suggesting that code should only be covered by copyright is stupid because there are limitless ways to express the very same concept in code and so the only thing that copyright protects is outright copying. I could take any piece of code, rewrite it and have it generate an identical result, and you'd never know I didn't write the thing from scratch.

So that said, I've come to believe that software patents in some form should exist.

The issue with software patents, as with patents in general, is that the process by which they are reviewed and granted is fundamentally flawed. Many patents are obvious, have obvious prior art, and are over broad. Also all patent holders that are not implementing a patent (the patent trolls) need to be forced to license the patent under FRAND terms.

That madness needs to stop. If we can't afford to effectively review patents then we should not be granting them.

And what do you think a patent does?


You shouldn't have to license "bounce back" and "pinch to zoom". It is like paying for the same work over and over because that one company was the first to patent it. These ideas didn't come from a vacuum; they were around for years before Apple used them.

Ok. 25 years. You may have heard of such a thing as mathematical logic? Thus, you are technically right, code is not mathematics. Code is mathematical logic and a bunch of other subdivisions of mathematics.


So yeah.... We are better off protecting general concepts, that don't bother disclosing the actual algorithms.

Also, there is no such thing as a software patent. We call it a software patent, because it references a process that is performed by a program code. All "software patents" are in their essence vague process patents.


I would agree with 3 points:
A) Full disclosure - source code mandatory!
B) Either copyright or patent, not both. The two deal with different types of IP.
C) Software patents don't need 10-20 years to recoup R&D costs, so 1.5 year protection term.(That was 2 years 9 months ago, that is how fast the industry changes)

Code == Logic. Math == Logic. Therefore, Code == Math.

*ALL* software is math.

Write some code that ain't, and SOMEONE can make a mathematical formula to describe it.

--The loon

I'm not sure how you down-voted. Even as someone who does not agree with software patents, you present a perfectly rational point on why the USPTO is so broken in general let alone with regard to technology.

Patents cover the underlying methodologies embodied in a given piece of software, or the function that the software is intended to serve, independent of the particular language or code that the software is written in. Copyright prevents the direct copying of some or all of a particular version of a given piece of software, but do not prevent other authors from writing their own embodiments of the underlying methodologies.

Software code is not "math".

Take Excel, for example. How was it created? It was constructed. It wasn't derived like a math formula. Nor is Excel some mathematical equation that existed in the abstract, just waiting to be discovered by a mathematical genius. Excel is a machine that was put together like any machine, only it's made out of bytes rather than steel/wood/plastics/etc. And machines (that is, the ideas behind them) certainly ARE patentable.

P.S.
What is the "you can't patent math" thing based on anyway? Why, for instance, would one be able to patent a chemical formula but not a mathematical one? Is it just because math is totally abstract and doesn't exist in "nature", but only exists in the minds of intelligent beings, and therefore nothing mathematical can be considered to be created/invented, but only discovered?

I ask because it seems that much the same could be said of chemical formulas. Chemistry, unlike math, does indeed concretely exist in nature, but I could see where one might argue that all chemical formulas already exist in the abstract, just waiting to be discovered, therefore chemical formlas can't be patented since they're discoverd rather than created/invented.

Seems arbitrary to say that math can't be patented, and I've not heard anyone actually rigoursly argue that proposition; I've merely heard the proposition asserted as an axiom as if there's no need to back it up at all.

Anyway, as I said above, I don't think the "math can't be patented" argument applies to software to begin with, since I don't see software as "math", but I do wonder about the whole "you can't patent math" thing to begin with. Seems like begging questions to me.

I only care from a philisophical standpoint. I don't really care in a real/practical sense. However, if a company spent billions of dollars on math research and discovered some mathematical formula that solved the world's energy problems, maybe that company would be justified in getting a time-limited patent to reap some return on their investment. Though it's likely that the discovery would be more along the lines of physics rather than purely absctract math. Maybe applied math should be patentable, but not purely useless abstract math? Or maybe math can't be patented, but the way one might apply it is patentable? :p