Subject: Re: patent trolls and X-licensors
From: simo <>
Date: Thu, 08 Jun 2006 08:13:43 -0400

On Wed, 2006-06-07 at 21:45 -0400, Taran Rampersad wrote:
> simo wrote:
> > On Wed, 2006-06-07 at 13:10 -0400, David H. Lynch Jr. wrote:
> >   
> >> simo wrote: 
> >>     
> >>> New understanding on the application of physics of signal transmission
> >>> should be patentable.
> >>> New understanding in mathematics should not. It's just all about the
> >>> field of the discoveries.
> >>>       
> >>     And what is the distinction between the two ? Are  you are arguing
> >> that physics is patentable, and mathematics is not ?
> >>     
> >
> > Not physics laws,
> > I said applied physics.
> >
> > You can't patent "the electromagnetic field", but you can patent a
> > transistor as it is a way to use electromagnetic fields that was not
> > thought before.
> >   
> Err. I believe you mean EMF, electromotive force. Electromagnetic fields 
> are generally considered bad when it comes to transistors as they induce 
> voltages in circuits. Transistors are semiconductors, and it's the 
> breakover that controls things. The electric motors/generators that 
> Tesla patented work with electromagnetic fields. But your point is well 
> made, maybe. He didn't patent the fields, he patented the results of his 
> experiments which proved his theories. The fact that his motors worked 
> was proof that his theories were sound. And he patented them. Of course, 
> he didn't do too well financially...
> So you can patent a working result of an experiment that proves a law of 
> nature. Which means that you can patent something which demonstrates a 
> law of nature, and perhaps even defines it within the scope of human 
> knowledge.

I think you are extending it beyond what I think is ok.
You forget that the invention must not be obvious (to someone skilled in
the art).

> Biocyte has a patent on all umblical cord cells from foetuses and 
> newborn babies.
> Genset (French) has an agreement with China to colect and patent DNA 
> from tribes in remote regions.

Actually I am against this kind of bio-patents as I am not ok with
patenting the human body, but this is another field and need a complete
different discussion, as the problems here are more on the ethic side,
even if economic problems and good science problems should be considered
here as well.

> John Moore's case is interesting, if you haven't heard of him:
> You can patent a piece of software that writes poetry.
> "...On Nov. 11, Mr. Kurzweil and John Keklak, an engineer, received 
> patent No. 6,647,395, covering what Mr. Kurzweil calls a cybernetic 
> poet. Essentially, it is software that allows a computer to create 
> poetry by imitating but not plagiarizing the styles and vocabularies of 
> human poets...."

This is clearly a software patent, physics is not involved at all!

> Bad enough that poets already have to die to make a profit...
> And there is software that designs circuits simply by evolving them, the 
> results of which are patentable.

This is another thing that I do not like that much, today circuits
should be awarded just a design patent (which is more like a sort of
copyright for electronic circuits than a classic patent) or perhaps
adapt copyright to cover them as well, as there is nothing really new in
most circuits being designed today using classical components.

> So, in theory you can't patent a law of nature... but you can patent 
> it's products and the evidence that the law itself exists. Tada!

Unfortunately today that's what the PTO, EPO and JPO allow around the
world, but we are actually criticizing these organisms in the first
place, as they are responsible of the fast dropping in quality and
usefulness of patents and introduction of patents on abstract fields
like software or business methods.