From: CurtAdams@aol.com
Date: Tue Dec 21 1999 - 19:45:33 MST
In a message dated 12/21/99 6:09:00 PM Pacific Standard Time,
bradbury@www.aeiveos.com writes:
> A quick query of IBM's patent network indicates 4987 patents on
> monoclonal antibodies. They were invented and many subsets
> were patented. What is the point?
The original invention was not patented.
>> Lots of innovation happens without patents.
>Of course. You have "copyrights", "trademarks" and "patents"
>in different areas. You have people who want to innovate and
>contribute those innovations to society or derive the benefits
>from being percieved as "good guys" -- witness the "old"
>Rockefeller foundation or the "new" Gates Foundation
>they are generally entirely divorced from copyrights,
>trademarks and patents.
Also, a lot is done to improve processes, and a lot is sheer luck.
Monoclonal antibodies are an example - Milstein was trying to
do research on antibody diversity and cooked up clonal antibody
lines. Then somebody in the lab realized what they had.
A lot of research comes out of academic work, which in the past
worked just fine without any need for patents and such.
>> Patents slow innovation markedly by a) increasing legal costs
>- yes, the legal costs are there, but there is *no* barrier
> to an individual patenting an invention himself, he must
> simply educate himself.
Ah, but every contentious issue leads to legal work and often lawsuits.
>> and c) creating complex negotiating requirements to address
>> all the possible infringements.
>That is the entire *point* of inventions.
We must be talking past each other here. When a company I worked
for invented a laser for angioplasty their goal was to treat sick people -
not to force anybody using a laser to negotiate with them. We did,
however, get torpedoed by a submarine patent. Somebody patented
the concept of using certain lasers to treat disease. Needless to say, this
"invention" was in no way necessary for our work - heck, we didn't
even know about it - but we spent a lot on lawyers and $50,000 to settle.
>I have two possible paths before me:
> 1) I can go work as a contract programmer for some XYZ company
> programming whatever they would like to have programmed.
> 2) I can *invent* things that may contribute to the productivity
> of society, patent those inventions and have companies & organizations
> coming to me to license those inventions.
>If you elimiminate patent/copyright/trademark law, you eliminate
>the motivation for (2). At that point I'm going to go and be a
>stupid contract programmer because there is nothing to prevent
>anyone from stealing anything I invent.
Yet a lot gets written under such circumstances anyway. There's
lots of open source software and freeware. It might stop you, but
it won't stop many other people.
The flip side is that without patents, it's far easier for somebody to
invent. With patents, many inventions require vast amounts of
patent research and infringement negotiation - which do nothing
to improve the product.
>> For an example of c), consider what will happen to software
>> innovation if the current trend to patenting algorithms continues.
>Patenting (previously generally known) algorithms is improper and
>those tests *should* fail in court.
But if current practices had been around, they would have been patented
when first invented and hence the patents *would* be valid. Also,
in the real practice of law, the better attorney often wins even if they
"should" lose. That's just reality. Certainly many crazy patents are
winning at least settlements. What's going to make that change?
>> We will soon be in a situation where every dozen lines of code
>> is a possible infringment (oops, used an XOR in those graphics).
>> If any sizable program requires negotiating on tens of thousands
>> of patents, software innovation will basically be impossible.
>Not entirely true. The semiconductor industry gets by with broad
>cross-licensing agreements.
Right; it does work for large corporations. Everybody else, however,
is completely shut out, including the small inventors you wanted to
protect in the first place.
>The fundamental question comes down to whether or not a patent
>is *novel*! That is a requirement for patenting. If it isn't
>novel then the patent should fall apart on a court apeal.
"Should". In any case, patenting novel techniques will create the
problem in a decade or two even if we prevent these obnoxious
"ex post facto" patents.
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:06:10 MST