Re: [Patents] You have rights.. Software Patents are Illegal

[Flemming Bjerke <flem@bjerke.dk>]

Fredag 14 maj 2004 14:39 wrote Jonas Maebe about Re: [Patents] You have 
rights.. Software Patents are Illegal:
> On 14 mei 2004, at 13:49, Flemming Bjerke wrote:
> > Well, I did not say that. I just said that in a computer, there is
> > nothing but
> > matter.
>
> In a human mind as well. That why looking at the way something is
> represented in a particular context cannot be used to judge whether or
> not it's patentable subject matter, because in that case everything is
> patentable subject matter. You would have to allow patents on thought
> patterns if they are new, non-obvious and if you could make money with
> it.
But, isn't all patents, patents on thought patterns? (Not that I would 
greive, 
if all patents were abolished.)  I mean, patents first become a huge problem 
when the symbolic expression of thoughts can "realise themselves" in the 
machine. i.e.  the computer (and when ideas, which only exhibits a confined 
understanding of the matter, can be realised and replicate themselves as 
living creatures (beyond our understanding) as in biopatents.)

>
> >> You have to see the distinction with the means
> >> with which the information is stored/processed, and the information
> >> itself (which is completely independent of those means).
> >
> > But, in a computer, there is no such thing as information in itself.
>
> "Information in itself" (as such? :) can never be in something, because
> information is non-corporeal and "something" is per definition
> corporeal.

Before you mentioned "thought patterns". Information is in a computer nothing 
but patterns of bits. Therefore, it can be copied. So, I agree while twisting 
your argument: In itself information is nothing. If there is nothing to 
order, there is no order (nor disorder). But, this argument applies only to 
abstract information, i.e. information which is handled without considering 
its meaning, as in computers. For humans, most information has a meaning, and 
the processing of it is dependent not only on its configuration, but also on 
its meaning. Furthermore, an important use of computers is to connect their 
abstract information processing with physical processes outside the computer, 
in a sense giving the abstract information meaning.

>
> > There is
> > only information which is stored or processed - and it is matter.
>
> And the means with which you process that matter is technology. That is
> completely independent from the information that this matter represents
> however. You cannot say that this matter processing is new or
> non-obvious, because it's not (the machine was built from the ground up
> to be able to do that and to be operated this way). 

But, as the processing of human ideas often depends on their meaning, 
patenting of ideas may be patenting of practices (practical ideas). 

> The only thing
> that's possibly new or non-obvious is the information that it
> represents. 

Since the new and non-obvious information may express new practices, it may 
be 
possible to patent it. The same is the case when computers are used to 
control new physical practices. (I think, this is what the EU-parliament 
accepted.) And you are even allowed to patent programs that control old 
practices.

> But information is not patentable subject matter.

Actually, you CAN patent abstract information that control processing of 
information. Unfortunately. 
>
> Just look at a book: all those letters on paper are clearly pure matter
> (ink). That doesn't mean that the story line the book is patentable
> subject matter 

I guess it is technically ;-) possible to make story-patents. Not that I 
think 
it is a good idea. 

But, I think you pinpoint an important argument: Patenting abstract 
information processing is quite different from patenting ideas, which for 
humans have practical meaning, beyond the very processing of information. 

As far as I have seen there are basically two kinds of software patents: 

1. Algoritme patents or program structure patents which are patens on how to 
process abstract information. 

2. Ideas of how to use programming, i.e. it may be described how programming 
is used, but often it is just made probable that some problem may be solved 
by programming (e.g. computer controlled bread baking).

The problem with the first sort of patents is that they are scientific 
discoveries which should not be patented because they as a rule are too 
widely applicable and their use are cogent.

The problem with the second is that they basically are not 
softwareinnovations. Often, they do not even offer a description of how to 
realise the idea, i.e. a description which shows what the invention consists 
of. That is, the only novelty is that computers are used to perform something 
that have not before (hopefully) been performed by computers. 

Of course, there are patents that combine 1 and 2, i.e. some algorithm or 
program structure is developed when solving some practical problem. Such 
patents is the least problemtic patents when they are bound to the practical 
context. But, as far as I can see Art 2a etc. of the EU-parliament allows not 
only such patents, but also 2 - if the problem is not just presenting, 
storing or processing information - but not 1. So even the EU-parliament 
decision will allow numerous trivial patents.

Flemming