🚀 go-pugleaf

> By the same line, all real objects are made of atoms, and since atoms,
> as natuaraly occuring objects are not patentable, no patent should be
> granted. 

A particular arrangement of atoms ceases to be a naturally occurring
object. However, a particular arrangement of mathematical algorithms
does not cease to be a mathematical algorithm.

> The patent office is not interested in the essense of things
> but in the practice and use of things. 

I do not believe this is the case. Historically business practices
have not been patentable, for instance, based on their essence as a
business practice, not based on their use.


> A new way of making, using etc something is patentable, even if the
> essense is not.

Maybe in other countries that have process patents. However, the US
has product patents; it is possible to patent _any_ way of making a
particular thing or _any_ way of achieving a particular effect can be
patented. Of course, I think this is bogus. I also think it is bogus
that independent invention is not a defense against a patent, if it
were, I would have much less objection to patenting programs or
theorems.

> Thus, although computers use math and may even
> essentially be math, the specific actions which the software performs
> is not.

Yes they are. Computation is computation.

> It is an invention to accomplish a certain purpose. It is that
> purpose, and that technique which are patentable. 

I especially object to a purpose rather thana technique being
patentable. 

> You could use exactly the same software to accomplish a different
> purpose, and you would not be in violation of the patent.

It depends on how broad the patent application is.

> Ie, it is the specific way in which those bits are pushed and the
> accomplishment of some purpose by that bit pushing which is
> patentable, not the bit pushing as an abstract idea itself.

You can't patent an application or purpose of a mathematical theorem
(AFAIK). I still do not see how software is distinct from mathematical
theorems or algorithms.

 - Maciej Stachowiak

-- 
guile -c '(define c call-with-current-continuation)(define(f s)(lambda(o)(let
((n(c(lambda(x)(o x)x))))(let p((n n)(l(string->list s)))(cond((not (null? l)
)(display(car l))(p(c n)(cdr l))))))))((f"Js nte ul akr")(f"utAohrGieHce\n"))'