🚀 go-pugleaf

In article <1991Aug31.184217.20959@aber.ac.uk> pcg@aber.ac.uk (Piercarlo Antonio Grandi) writes:

   dak@sq.sq.com (David A Keldsen) writes:

   > That is disputable.  In particular, part of the 5th amendment protects
   > property rights, and patents have a long legal history of being 
   > considered property rights.

   No, they have never been considered property rights. You can have property
   rights to a patent, but the patent does not confer on you any property
   rights to your invention, just a monopoly of some sort. So the 5th amendment
   protects ownership of a patent, not of the invention described therein.

   Under the US Constitution expressions and inventions are not considered
   personal property, that's why there is an explicit provision for the
   possibility of granting of monopolies to authors and inventors.

True.  In fact the defense of patents that goes "I believe in protecting
intellectual property" really is a defense of the abolishment of all
patents.  The patent system serves to RESTRICT intellectual property.

In case it isn't clear how this is the case, I'll offer an analogy.  You
decide to build a house, sit down and design it, order the materials, and
build it.  It is your property, of course.  Then one day, a year or so after
you've lived in it, somebody comes along and says you have to pay them
$50,000 per year to continue living in the house or else destroy it and build
another, different house (or at least change it significantly), and even
then you still owe them $50K for the first year you lived in it.

The reasoning?  They happened to have designed and built a similar house
one month before you did, and filed the design with the "Bureau of
Housing Property".

The correlation with patents is simple: if you come up with an idea,
implement it, and use it in a typical way (i.e. sell the resulting
product), you don't really own your own intellectual property if it turns
out someone has done basically the same thing (come up with the same idea)
before you did, or at least filed it before you did.

Any true believer in intellectual property would have to be totally against
patents of all kinds (unless they had a reasonably solid way of
distinguishing someone stealing an idea from someone who just happened to
come up with it on their own, as happens often enough, so they could
still award such people the right to their own intellectual property while
still denying stealers access to others' intellectual property).  Very few
people feel that strongly about intellectual property.  Some of us,
however, see computer software and software programs primarily as
extensions of our own abstract thinking and ideas, ways of consolidating
those thought-patterns in a manner that makes them more reliable and easier
to interconnect.  I, for one, see it this way.  So when I say I want to
protect my rights to intellectual property, I mean I want to protect my
rights to implementing my own ideas in a realm that I believe is entirely
one of ideas -- software.  That is why I'm against software patents.

The primary reason for patents to exist is to encourage the publishing of
inventions.  Every software patent I've heard of or, so far, can conceive
of, is inherently published when its program is published.  In any case,
the percentage of inherently published software patents vs. all possible
software patents must be hugely higher than in most other fields -- for
example, in chemistry, delivery of refined oil offers little in the way
of information useful towards determining how to refine oil.  There are
specific cases in other fields where publishing would be an automatic
outcome given the sale of any device incorporating the patentable technique,
yet the patent office still issues patents for such things -- by the
same reasoning, I feel it is legitimate to ask for the abolishment of
all software patents even though there may well be a few cases where
publishing isn't automatic upon shipping a product, because in the huge
majority of cases, it is.  (In fact, as we all know, publishing the
idea often happens before the idea is patented as a software patent,
and in many other cases, it isn't published because it is so obvious, as
in the case of XOR cursors, backing store, and using a canonical byte
ordering in a binary file.)

Admittedly the term "software patent" needs a better legal definition.
One approach to finding one that I could use is whether human thought
processes can substitute for the complete process described in the
patent.  If so, it is a software patent.

For example, if someone hands me unrefined oil and says "refine it", all
the genius in the world won't allow me simply to speak, write, or otherwise
communicate refined oil as output.  Similarly, being handed a bunch of
chemicals won't allow me to make aspirin, and so on.  Instead, while I must
use intellectual processes, I would have to resort to mechanical techniques
to accomplish the job.  Thus, it is not a software patent.

However, if someone shows me a listing of C code and says "compile it", I
might well be able to write down the output of that code as, say, 68K or
i486 assembler, and in the process use, in effect, many currently patented
techniques, such as IBM's register coloring algorithm, that I may have made
up myself.  Note that just because I use a mechanical technique -- pencil on
paper -- does not invalidate the premise, because the paper-scratching is
NOT the true output of the process, nor is the original listing of C code
the true input -- the listing _represents_ the input, as does my output,
whether these are in verbal, visual, or other form (braille, for example).

Also realize it is impossible and certainly not desirable to try to put an
upper bound what the human mind is capable of.  No patent should be issued
on an algorithm to rapidly find the cube root of a 100-digit decimal number,
because that would be a software patent.  One might justify such an
attempt by saying "the human mind is incapable of performing this task in
any reasonable time frame", but that argument must not be permitted, and in
this particular case, I believe I've seen proof that at least one person
(the name "Dr. Kim" comes to mind) can indeed do it, possibly faster than
even today's computers.  So even though I can't possibly compile your
average C function into assembler as fast as a top-notch compiler running
on a computer, that doesn't mean no human can do it, so task size and
complexity never should factor into whether an idea is patentable.  (This
is not the same as say obviousness shouldn't factor into it -- what
distinguishes obviousness from size or complexity is the concept of
inspiration.)

In summary, we should go back to the 1981-and-earlier standards for
patentability, and abolish software patents retroactively.  Processes that
include software may still be patented, but use of the software (equations)
that is in the patent should not, by itself, ever constitute infringement.
(In other words, still permit patenting an overall process that happens
to involve a computer and a running, specialized program, but don't
prevent anyone else from using the special algorithms in that program when
they write their own.  This was the original change to patentability that
unfortunately opened the floodgates on pure software patents -- up until then,
from what I understand, any process that included a software program was
not patentable, and in my opinion that shouldn't be the case.)

How do these thoughts sound to the various net folks out there?
--

James Craig Burley, Software Craftsperson    burley@gnu.ai.mit.edu