Wouter van Ooijen wrote:

>>> - giving a limited-time monopoly encourages investment in new
>>> discoveries that otherwise would not be discovered
>> 
>> This is a prerogative that is not proven at all. Same thing with
>> copyright on music, for example. There is nothing that proves that
>> there won't be music without copyright on it. Actually, history
>> shows that there was music without copyright. And there were
>> inventions without patents. Things would be different without them,
>> that's for sure, but who's to know how exactly? And whether that
>> would be worse?
> 
> That is a viewpoint, but IMO it is equally unproven that patents do
> not encourage new developments.

Right, I agree. If it's not proven either way, let's take the argument
out of the equation. This is why I wrote that comment. And with this
argument out of the equation, there is not much left in favor of a
patent system.


> I think patents will not disappear altogether, so we can at least agree 
> that on what should definitely not be patentable, like things that

If those are necessary for patents to make sense (to you), they won't
make sense (to you) for a long time -- if ever :)

> - are obvious

It is difficult to define (court-safe) what is "obvious". What's obvious
to you doesn't have to be to me. Which brings us back to the several
100k EUR or USD or whatever that may be necessary to defend your view of
"(non-)obvious". Not being obvious is already a necessary criterion in
most if not all jurisdictions and it doesn't help.

> - can be invented with (very) little investment

It can be difficult to determine how much investment was necessary to
invent something and it is tricky to define how much is "(very) little".
A nifty little mechanism that solves a niche problem and hasn't been
there before could maybe have been invented in an hour late at night in
a bar. But maybe it was invented by one who has worked in the field for
decades and was intimately familiar with everything involved. How do you
calculate the necessary investment? A pencil, two napkins, an hour and
three beers? Decades of hard work of becoming familiar with the issues?
How do you compare it to the set minimum? Court-safe?

> - are just there, waiting to be read (genome, geography of pluto)

In a way, at least in the way some philosophers see it, everything is
"just there, waiting to be read" :)  If you didn't mean the
philosophical approach here, how do you define "just there" (again
court-safe)? Things that are "just there" can't generally be patented
under the current rules, and it doesn't help. OTOH, a specifically
isolated genome is not that different from a specific alloy, and there
may be some novelty. 

> - take a large investment, but will be invented anyway

I don't understand this one at all.

The problem is that when it gets down to define rules that are supposed
to make the system work, all kinds of assumptions need to be made and
things get so complicated that it seems to me that there just isn't a
set of rules that makes sense and has a chance to work. Example for this
is that two of the rules that you wrote are already in place -- and
don't help at all.


>> Which basically gets down to that patents in principle don't work
>> (anymore).
> 
> For the current practice I agree. That does not imply that patents
> can't work at all.

Maybe they can, but even if they can, this still doesn't answer the
question whether a working patent system is better than none at all. 


>> That's just a temporary solution, until the 10k (10k what? Euros? :)
>> brings the same level of uncertainty as the 1k now.
> 
> 10k (or maybe 50k) should allow a patent officer to do a decent
> search. This figure should of course be indexed.

This is not just a question of inflation. Indexed to what exactly? The
amount of public knowledge? How to measure this? Patent officers can't
just search patents; they'd have to search everything published. This
grows with every year.


>> if you don't have the several 100k that it probably costs you to
>> defend the patent, it may not even be worth the 10k.
> 
> That's why I argue that the main cost should be up-front, not in
> later procedures.

No matter how much you put in up-front, there's still the possibility of
a court challenge. Your patent can be 100% safe, but you still may need
to shell out the money to defend it.

Gerhard
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist