Re: [Patents] Software is not hardware, social sciences are not natural sciences.

[Russell McOrmond <russell@flora.ca>]

On Wed, 10 Mar 2004, Gregory Aharonian wrote:

> >When I say software is not a technology I simply mean software is not
> >a technology.
> 
> You say nothing here.

  There are none so deaf as those who have a financial incentive not to 
hear.

>  I see the word "technology" and look in my dictionary, and see the
> definition "application of a science".

  So an application of politics which is a social science, is a
technology?  Do recognize no differentiation between natural sciences and
social sciences?

http://dictionary.reference.com/search?q=technology
http://dictionary.reference.com/search?q=natural+science
http://dictionary.reference.com/search?q=social+science

  Do we have to update dictionaries to clarify what should be obvious 
which is that technology is the application of natural sciences, not 
social sciences?

> Software is an application of the sciences of mathematics and physics.

  That is your definition.  It is not mine.

  Software is the rules that govern the operations of a computer.  If we
are to make comparisons it is far closer to political science than
physics.  In fact I am saying that software has absolutely nothing to do
with physics.

  I recognize that some "software engineers" will want to disagree, but
software rightfully belongs in the social sciences.

  That being said, patent policy is economic public policy and should not 
be expanded to new subject matter types simply because someone believes 
they are a technology.

> The dictionary says nothing about forces of nature, physical objects,
> etc. - just that technology/technical is in the realm of applied science.
> Much like digital hardware is an application of the sciences of
> mathematics and physics.

  So, any science?  Is the process by which we elect politicians a 
technology?  Should we be granting 20 year monopolies if someone comes up 
with a "useful, novel, unobvious" way to elect politicians?

  Then again, patent law is itself a technology given it is "economic"  
public policy (economics is a social science) encoded in law by
governments (political science is a social science).

  Can you tell me where you draw the line in expanding the concept of 
technology to social sciences?  What in our lives is not a technology in 
your mind?

  Do you honestly not recognize the absurdity of what you are asking
people to believe?  You are suggesting that any "novel and unobvious"
application of any type of science (social or natural) should
automatically be granted statutory monopolies via patent law?

  We can't even get into the question of "industrial applicability" as I 
suspect you believe that there is nothing that doesn't have "industrial 
applicability" given you likely believe that the only model of economic 
production is industrial production.  Likely the "useful" test to you is 
just a waste of space in patent acts, as is the "statutory" test given you 
believe in unlimited patentability.

  Likely if you got your way you would get rid of the "unobvious" test as
well given you only ever talk about the "novel" test, as if it was the
only legitimate requirement for patentability.

> So if you are going to allow a patent on a specific application of
> mathematics and physics in the form of a specific digital circuit,
> given that you can convert the specific digital circuit into a specific
> computer program, there is no reason not to treat them both as
> equally patentable.

  Before the printing press people had to "hand make" each book.  After 
the printing press we could then "mass produce" books without each book 
having to be "manufactured" by hand.

  So, given the printing press is considered an invention, and each book
that you may print is a unique thing that was previously created
individually, your logic suggests that each book should be patentable?


  I see no logic to your argument that because a manufacturing process is
patentable that if we come up with a way (technology or otherwise) to
achieve the same goal without per-design manufacturing that the end result
should still be per-design patentable.


  Patent policy is economic public policy, and needs to be analyzed that
way. People are moving more and more of the complexity of computing from
hardware into "software" (using it as broad as you mean it, which is
different than how I use the word) to save money.  This gets right to the
roots of the issue (ignoring for the moment the absurdity of the "social
sciences are natural sciences" argument you present) which is that with
the lower cost of development then comes a lower necessity for large
capital investment and thus a reduction in the justification for a long
statutory monopoly.

  There may be an argument for a shorter term and much more limited sui
generis protection for designs of the type you speak about, but I see no
logical argument in favor of patentability.

> That is the message of US caselaw - applications of software algorithms
> are patentable.  There is absolutely no logic to deny such specific
> applications the possibility of patenting.

  You may not recognize the logic, but that doesn't mean it isn't there 
for the rest of us.  Your equating of social sciences to physical 
sciences, suggesting that politicians are really just engineers, is what 
lacks logic.

  The message of US caselaw is that the US parliament did not clarify
patentability enough to stop the abuse of the law by patent lawyers,
monopolist executives and patent paralegals like yourself to expand the
law into areas where it simply does not make sound economic public policy.
Like other forms of voodoo economics like what created the .COM bubble,
this flaw needs to be corrected to reduce ongoing economic harm.


  I find discussing this with you quite useful as it allows me to think
more about some of the arguments, and better articulate to
parliamentarians just what type of abuse will be tried.  It allows me to
make better suggestions about how we need to fix patent laws to clarify
this situation.  This is a case where older laws need to be modernized to
take into consideration recent arguments that could not have been imagined
at the time that these laws were first drafted.

  Then again, I've never had to discuss the difference between a social
science and a natural science to a politician so we are further ahead
with them than the discussion here with you :-)

> What you are complaining about is overall broad software patent claims
> that look less like specific applications.

  That is *your* complaint, not mine.  You believe that patent quality is
the only problem.  While I agree with you that this is a serious problem,
we disagree that it is the only problem.

  How about if you devise a way to get 95%+ patent quality by your own
standards.  I believe that even in your world it would be reasonable to
put a moratorium on software patentability until it can be proven that
high quality patents in that specific subject matter can be achieved.  I
believe that for any type of patent (even those of a subject matter type I
believe should be patentable) that is it better to given inadequate
protection to a good patent than over-protect an invalid one.  The norm in
our economy should be free market competition, not government granted and
protected monopolies.

  I realize this may seem too "right wing" to some, but misapplied
economic public policy can be very harmful to the economy. The default
should be for governments to be "hands off" free markets unless a properly
researched and defined public policy need is articulated by governments
(not lawyers or patent offices, and certainly not other pro-patent
economic special interests).

-- 
 Russell McOrmond, Internet Consultant: <http://www.flora.ca/> 
 "Make it legal: don't litigate, use creative licensing" campaign.
 A modern answer to P2P: http://www.flora.ca/makelegal200403.shtml
 Canadian File-sharing Legal Information Network http://www.canfli.org/