Re: [Patents] Cryptographic patent... and research.

[Russell McOrmond <russell@flora.ca>]

On Mon, 14 Jul 2003, Jozef Halbersztadt wrote:

> Hartmut Pilch wrote:
> 
> > 1. 90-95% of all patents granted are valid according to PTO criteria
> >
> Where can I find PTO criteria?

  Each patent office has their own criteria, based on domestic 
legislation, examination guidelines, case history (both domestic and 
otherwise).

  - Whether something is statutory is partly what will be defined by these
tests.  Once a subject matter is deemed unpatentable by statutory
exemptions, there shouldn't be examination error at all.

  Patent examination guidelines will need to be written to ensure that
non-statutory subject matter patent applications are rejected, not written
to encourage patent lawyers to word applications such that the
unpatentable appears to be patentable.  There needs to be some incentive
for patent offices to recognize that their constituency is the general
public, not patent filers.  Their job is *not* to help patent filers file
for patents -- their main job should be to examine patent applications
specifically looking for adequate justifications to reject them.

  - Whether something is novel is going to be testable by applicability of
prior art.  In software it is unreasonable to have a patent that lists no
non-patent prior art given the huge international library of software --
most of which is created outside of a patent context, and much of which is
available online.

  - Whether something is unobvious requires more testing than happens 
currently, and this will be one of the harder criteria to adequately 
measure.

  - Useful?  I'd like to see some clarification on this.  Whether
entertaining a cat with the non-visible (to what?  the cat?) light from a
laser-pointer is any more 'useful' than a perpetual motion machine isn't
worthy of a debate.

> > 2. 90-95% of all patents granted represent substantial investments
> >    which are difficult to recoup without a 20 year monopoly.

  This one is also hard.  It does go back to the question -- why should 
the costs/inflation of patents only be born by those skilled in a given 
subject matter?  If it is hard for economists to give adequate economic 
analysis of subject matter to guarantee the suggestion above, then it 
should be equally hard to justify the existence of patents on that subject 
matter.



  I suspect that if this analysis was done that the breakdown of 
patentable/non-patentable would be similar to the mind-matter breakdown 
suggested elsewhere.  I guess I wanted to clarify that I don't think that 
mind-matter test is inadequate, but that it is not something that will get 
policy makers excited enough about the issue to deal with it.

  Give them incentive to do their own studies to prove the patent system
is failing in critical subject matter and you will find that they end up
on our side of the debate.  Eventually you will end up only with the
extremists and protectionist special interests remaining on the other
side, and pushing for protectionist policy is politically unlikely given
the pro-trade/pro-competition policies generally pushed by western
governments.


  I guess my only real input into this forum (which has been a great
resource in all my research) is to suggest that the people in this forum
try to think as politically as they can.  We need to have multiple angles
happening at the same time:  responses to particular proposals, proposals
of our own, and attempts to excite policy makers of all types to get
involved to help fix the policy problems in this area.

---
 Russell McOrmond, Internet Consultant: <http://www.flora.ca/> 
 Governance software that controls ICT, automates government policy, or
 electronically counts votes, shouldn't be bought any more than 
 politicians should be bought.  -- http://www.flora.ca/russell/