[Patents] Re: Patents digest, Vol 1 #299 - 7 msgs

[Edward Avis epa98@doc.ic.ac.uk]

On Mon, 16 Oct 2000 patents-request@liberte.aful.org wrote:

>From: Gregory Aharonian <srctran@world.std.com>

>>So I could protect my program, which is an implementation in C of a
>>particular sorting algorithm.  But I couldn't stop others from making
>>their own inventions based on the same algorithm.  This is exactly what
>>copyright on computer programs achieves.
>
>And that is why people are increasingly abandoning copyright protection
>for software - because it gives no protection,

I'm afraid that what you call 'no protection', I would call just the
right amount of protection.  Not too much for an open and competitive
market, and not too little to encourage development.

>and turning to patent protection.

Of course they are, if it is available.  If Electrolux had the chance
of trademarking the phrase 'washing machine' they would do so without
blinking.  Ordinary trademark law would be feeble in comparison.  But
this does not mean that the status quo is the wrong balance between the
interests of big producers, small producers, and consumers.

>If I invent a new sorting algorithm, and can't control how
>others exploit it with copyright protection, I will either get a patent
>or not disclose my sorting algorithm

Could you name a sorting algorithm for which this has happened?  As I
recall, they all seem to have been discovered despite having 'no
protection'.  It is just possible that there is a secret collection of
undisclosed sorting algorithms in use exclusively within some large
company...

>Increasingly copyright is becoming absolutely useless to protect
>software

I see a lot of companies making a lot of money based on software
copyright.  How is it useless?

>- it is not an alternative to patent protection.

>From the company's point of view, getting a patent they can use to
hinder competitors has to be a good thing.  But as I've said, it does
not follow that granting these patents is the best course for the market
as a whole.

>This is part of the reason for the explosion of software patents in
>the US,

The reason companies file for software patents is 'because we can'.  You
can't go from the fact that large numbers of swpats have been granted to
the conclusion that they are 'useful', 'essential' or even 'a good
idea'.  All you can say is: they exist.  Then you have to carefully
analyse whether they are having a generally beneficial or generally
negative effect - which I don't notice you doing.  Can we have an
argument other than 'companies apply for them, so they must be good'?

>Better to fix the patent examination process, a solvable problem, that
>trying to convince people to give up patents.

Fixing the patent examination process would eliminate stupid patents
like one-click and 3d pie charts, which has to be a good thing.  But
even if you could weed out all the silly patents - things which
were not new, things which were previously considered too obvious to
patent, generalizations from a single algorithm to a whole family most
of which have not been discovered, patents which effectively cover a
problem rather than a single solution - even then, I don't think that
software patents would be a net benefit to the market and to
society.

And, BTW, I see no sign that the examination process is getting fixed.  
AFAIK the EPO board did not vote to fix the examination process first,
and legalize software patents later.  Quite the opposite in fact.

>(And Hartmut, stop calling EPO types "evil" and other insulting phrases.
>It makes it that much easier for them to ignore all of your arguments.)

This I have to agree with.  Even if they are evil :-)

>>Unless you want to allow mathematics to be patentable also.  That may
>>happen sooner than you think; since a computer program is equivalent to
>>a mathematical proof, patentability for one would lead to patentability
>>for the other.  I can just imagine all the patent lobbyists complaining
>>of the inconsistency that a computer program can be patented but pure
>>mathematics cannot...
>
>And they should complain - it is inconsistent.

Yikes!  I thought the claim that software patents could lead to patents
on mathematics might have just been scaremongering by anti-swpat
lobbyists.  But obviously not!

>And not only for software.  Increasingly there are many mechanical and
>hardware designs that are equivalent to mathematical proofs (if you
>are knowledgeable about the latest AI mechanical CAD systems - which
>apparently many of you aren't).

If you design a machine, you get a patent on it so nobody can copy its
design.  But surely, somebody else could sit down at the same computer
and use the same mathematical principles to come up with a different
design which would not infringe the patent?

Could you explain what you mean?

>So if you reject software patents because some of them are logically
>equivalent to a mathemtical proof,

No, that's not why I reject software patents.  I reject them on economic
grounds.  I see plenty of evidence that they create a legal minefield
for small developers and stifle competition, and I see little evidence
that patent monopolies are necessary to reward developers.  Even if that
were true in some cases, you still have to consider the net effect on
the market as a whole, and weigh up the instances in which swpats are
beneficial against those where they are harmful.  I believe the latter
group far outweighs the former.

Logical equivalence to a mathematical proof, like pointing out the
obvious fact that a computer program is fundamentally different to a
tangible object, is a philosophical reason to not grant patents on
software.  It is a clear principle which can be used in law to define
patentability - a physical object can be patented, while something which
is purely information is subject to copyright.  It's a way to make the
patent system serve the best interests of society while still having it
be consistent.

>then you have to reject mechanical and hardware patents as well
>because some of them are logically equivalent to a mathematical proof
>as well.

I doubt that can be the case in quite the same way as for a computer
program, but whether or not it is, it doesn't follow that you have to
reject patents on mechanical devices.  Just to make things clear:

Patents on physical objects are on the whole economically beneficial.

Patents on software programs are on the whole economically harmful.

The patent system is not an end in itself.

This last point is the most important.  All the patent offices in the
world exist only because governments have decided that it is
economically useful to grant time-limited monopolies for certain classes
of invention.  Any considerations about consistency of the patent
system, what building the patent office is located in, and what size
paper patent applications should be submitted on, are secondary.  What
matters is the net benefit to society (which includes economic effects
and personal freedom).

That is not to say that the law should not be clear on what is
patentable and what is not.  But there's no reason why including or
excluding a particular category from the scope of patentability should
create legal problems.  For example, if there were serious evidence that
farmers were being harmed by patents on agricultural machinery, and
evidence that most such machinery would be developed anyway without
patent protection, I would not hesitate to add a special exemption for
this class of object.  (Of course, I don't think the situation really is
like this for farm machinery.)

The distinction between software and hardware is fundamental and much
easier to see than any difference between a tractor and a lorry.  It is
not at all difficult for patent offices to accept one and not the other
as suitable material to be covered by a patent.  Over the past few years
the EPO has tried hard to muddy the waters with a meaningless
distinction between 'software' and 'software not as such', but this
doesn't mean it couldn't exclude software from patentability if it
wanted to.

The scope of patents should be decided by elected politicians deciding
what is socially beneficial and what is socially harmful - not by patent
offices worrying about what is consistent and what is inconsistent.

-- 
Ed Avis
epa98@doc.ic.ac.uk