|
Authored by: Anonymous on Tuesday, July 24 2012 @ 12:23 PM EDT |
I picked, from among a few definitions in a few
dictionaries, a definition that was simple, broad, and not
tagged as archaic:
MACHINE: " a mechanically, electrically, or electronically
operated device for performing a task"
You have to concede that "CPU + program" meets the above
definition of a machine. Part of the argument is about
whether it's "new".
(To a lawyer, that argument is fundamentally about Section
103, not Section 101.)
If performing a task is part of the definition, it's not
completely crazy to believe that new software for a
different task makes a new, different machine.
That's not really a "policy" belief, but it's one that's
very appealing to lawyers. The underlying policy is to
"protect" new, important, difficult inventions (anything
that requires "skill in the art") by extending patents to
them - subject only to vaguely-delimited countervailing
principles that certain things aren't patentable because
patenting them would be tantamount to patenting thought or
speech.
But I ignored the rest of the argument about whether
software gives rise to a "new machine". Sure, "CPU +
program" is a machine, but what's being claimed in the
patent? Where's the invention? Strictly in the software.
Therefore, shouldn't we require that the software in
isolation meet the definition of a machine? (Courts
disagree about the answer to my "should" question. It's
possible to read the patent act either way; the Fed Circuit
says "no", but the Supreme Court strongly hints "yes".)
Let's try looking at the software in isolation and see where
it leads: is software a machine?
Is software "an electronically operated device?" I'm not
sure. I do think it's a "device" (I looked up that
definition too) but I can see arguments for and against
"electronically operated." It's a judgement call, and
policy considerations will sway a judge's interpretation of
language. Empirical evidence that software patents are
harmful will be useful.
Remember too that judges are used to "legal fiction". If
declaring that software is a "machine" for purposes of
patent law is the only way to uphold softare patents, they
are likely to do so. Judges aren't good at deciding
technical facts, and they know it. They'll defer to the
USPTO on the assumption that those guys know what they're
doing. (Speaking of legal fictions, patents have a
"presumption of validity.") Since the USPTO has issued tens
of thousands of software patents, courts will assume that
there is some innovation going on that needs protecting.
That would change if the USPTO showed up in court and said
"these stupid software patents are overwhelming us!" but
that's not likely, due to the fact that the USPTO gets paid
on a per-patent basis. What may eventually happen is that
other industries will complain that software patents are
harming them. But it's an indirect harm, and so industry
will be slow to react and courts will be slow to listen when
these other industries try to "interfere". Luckily, the
software industry by itself is big enough that Congress and
the Supreme Court are finally paying a little bit of
attention.[ Reply to This | Parent | # ]
|
|
|
|
|