Re: [Patents] pooling patents for progress and protection

[Peter Vandenabeele <peter.vandenabeele@mind.be>]

On Fri, Jan 03, 2003 at 02:46:38PM -0500, Gregory Aharonian wrote:
> >I don't know the correct way for fighting software patents.

Please clarify who made the above statement ? Since you address your mail 
to me (the To: line in your mail), it seems that you suggest I said that ? 
I checked my mail to which you replied and this statement is not in that
mail. 

May I urgently request to use correct quoting techniques in mailing list 
discussions. I do not like my statements been modified by the person 
replying to me. This was also the case in the previous mail you sent to me.

> Easy solution - help patent offices give more time and prior art resources
> to their examiners so that the patent examiners can issue far fewer but a
> much higher quality software patent. 

I read somewhere that patent examiners are rewarded for the amount of patents
they grant, not for the amount of patents they critically reject ? Could you
please explain if there is any rewarding system or quota for accepting or 
rejecting a minimum % of applications ?

> I have long argued that over 60% of
> thes issued software patents are invalid.  Most examiners are smart enough
> to argue so against patent applicants, if the examiners were given more time
> and money and resources to do their jobs.  And most of the software
> patents everyone whines about are in this 60% of invalid patents.

There are two levels here:

- trivial and non-novel patents are a major problem, and that problem could
  indeed be reduced by spending even more budget on that. However, with the
  growth and world-wide extent of the SW business (due to the very low barrier
  to entry for SW development), it is simply impossible to check against 
  all SW implementations in a certain field. This will only become worse as
  more and more is done in software. Eventually the system will collapse if
  no change is made.

- the fundamental problem is that a general/monopolistic protection of "ideas"
  just equals to a political decision to make it impossible to create a system
  of Free Software that is guaranteed to remain Free (as it was possible with
  the GPL Copyright system). Allowing Software patents (and other patents on 
  "ideas") is simply disallowing people the fundamental right to build up a 
  Free Software system. People are forced to enter into legal constructs
  to protect their work with similar, complex and detestable schemes (and with
  even more money flowing to patent lawyers instead of to actual software
  development). It is not because I describe such schemes, since they will 
  eventually have to happen in the countries that accept SW patents, that I 
  do not detest that we have to loose our time on this, instead of just 
  continuing our work.

> And much can be said about hardware patents, optics patents,and patents in
> many other areas of technology - patent offices issue to many invalid and
> overly broad patents.  Yet one more feature software shares with other
> areas of technology, and one more reason software should be as patentable
> as any other technology.

In optics, in car manufacturing, in ... [essentially _industrial_ processes],
the barrier to entry is so much bigger, that there are a limited number of 
large (and novel but financed or backed by Universities) players that compete 
with each other and it is still somewhat realistic to get an overview of 
who is doing what. Also, since the result is of a mechanical nature and not
in a "text" form, it is easy to copy by looking at the implementation, so 
secrecy is a less efficient method there.

In software, since there is no barrier to entry, except your personal 
capability
to write good and clever programs (and access to some low-end computer and
networking that you can get at schools, libraries, ...), the number of 
eco-systems
just grows astronomically. Every individual behind a computer can be writing
software and adding to the general available amount of software. The number of
eco-systems grows in the millions. It is just not practical to use a system 
that 
is interdependent on the activitiy in the whole world in this field. You must 
have a system that allows independent eco-systems to co-exist next to each 
other, 
with an option to allow _positive_ interdependency (with Copyright you can 
_grant_ a right to use a program to another eco-system, but you cannot 
_forbid_ 
another eco-system from using their own implementation).

Trying to extent the protection of what is "invented" in one eco-system to
allow to forbid usage of that basic method in all other eco-systems is just
not efficient.It is too strong protection for one entity "inventing" 
something 
and blocking all other entities, especially as the number of eco-systems 
becomes
increasingly large in SW development. Because it is world-side interdependent,
this does not scale. 

For software, it is much more efficient to use Copyright and Secrecy as a 
protection 
mechanism on one possible implementation that is developed in parallel 
eco-sytems. 
These still allows _positive_ granting (through a license and through handing 
over 
a copy of the source code under NDA) and this perfectly allows an honest 
business 
based on proprietary software to be built on this model (of secrecy and 
licensing), 
without blocking others to run a parallel Free Software eco-system (and 
actually
the proprietar business has one major advantage over the Free SW business, 
that he
can invest once in writing a piece of SW and then sell the same software 10, 
100,
1,000 or 1,000,000 times, while Free SW under GPL can only sell it ai limited 
number
of times until someone decides to publish the source code he has received).

> With regards to the patent pool idea, in Europe it won't have any impact.
> Europe already has a powerful deterent to patent abuse - that losers pay
> in patent litigation. 

Which even further deters the smaller players that are under attack ! I will
not attack a large top 10 SW company (and even I would be in my right, the 
costs
would be peanuts for them), but if they sue me (and that is the whole 
problem),
if I don't have my case well defended, I may have to pay even more when I have
lost the case.

> This stops many bad patents, software and non-software
> from being asserted in Europe.

Please explain how you see this ?

> As a practical business matter (and I
> realize few of you here know much about running real businesses),

I run a real business in Free Software, we just extended to the USA and we are
making profit. And again, SW patents are the main threat to killing my 
business,
not the way I run my business.

> you
> shouldn't worry about having software patents and business patents in
> Europe. 

I do worry a lot about that.

> But if you hear that politicans plan to change the "loser pays"
> aspect of patent litigation, then worry very very much.

See above.

> Heck, if you actually want to do something useful (I am doubtful), you
> ask each of the 100,000 people who signed the anti-software patent
> petition to donate one book on software to either the EPO or PTO.
> Having access to 100,000 such books would provide the examiners with
> ample amounts of prior art to deny many software patent applications.

How would you ever search through these books if time is already that limited 
? 
And software development is not in books, but in mailing lists, in privately 
developed software (of which the code is not published, ...), in CVS trees, 
in the n*1,000 programs in the commercial and non-commercial distro's, a
program some M.Sc. student in Thailand wrote as part of his thesis ... The
amount of Sw that is developer is just so increadibly vast that you cannot 
even begin to claim that you have a 90% certainty that a certain "invention" 
is really novel.

Sincerely,

Peter

> Greg Aharonian