August 10, 2002

Commentary: On IP, or intellectual problems

-By John Fitzgibbon -

Though Open-Source more or less eschews the use of patents, the
impact of the use (or abuse) of patent law by commercial interests is an
issue that resonates across the software development community.
In essence, a patent establishes a legal monopoly on an invention for 20
years.

The patent was "invented" (sorry, couldn't resist) to protect and encourage
inventors. One could picture the scatty professor racing from the tool shed to
the patent office to stake a claim on the next technological marvel, thereby
ensuring his or her fame and fortune. These days, most inventions occur as
part of "just another day's work," for which the inventor, (or, more likely,
group of inventors), gets "just another day's pay." The only person racing to
the patent office (metaphorically speaking, of course) is the corporate
patent lawyer. The inventor(s) may still get the fame, but the bulk of the
fortune almost invariably ends up on a corporate balance sheet (that is, of
course, assuming the accounting is above board).

In and of itself, this change in emphasis is not a bad thing. Many inventions
require multi-million dollar investments and facilities that go way beyond the
scope of the average tool shed. What incentive would there be for Merck to
invest billions researching new drugs if Pfizer could just sit back and wait,
then knock off cheap copies?

But in the murky waters of software technology and associated intellectual
property (IP), current patent laws show their weaknesses, and can hinder
innovation rather than promoting it.

IP "land grabbing",
where an individual or company buys patents that have not been pursued, then
seeks to profit by vigorously upholding the patent rights, is morally
reprehensible. Deliberately allowing a genuinely held patent to creep into
general usage before exercising the patent rights is equally reprehensible.

But these are pretty straightforward issues -- they're simply wrong, and should
be dealt with legislatively. A trickier question to address in the software
world is this: "What exactly constitutes a patentable invention?"

There are two principle trains of thought on this subject. One holds that
software should be treated in the same way as any physical invention, and
therefore is patentable if it meets the general patent requirements of being
"new, useful and non-obvious." The other train of thought holds that software
is more like literature, and should only be protected by copyright, rather
than patent. Both sides generally agree that source code itself is creative,
and therefore copyrightable, (or "copyleftable," if the author so chooses).
The disagreement centers round the intangible "thing" that the source code
represents. Is this "thing" (i.e. the software) patentable?

Ideologically, I favor the second argument, which holds that it is not. Even
putting ideology aside, the practical issue of interpreting the definition of
the "thing" that a piece of software represents has led to overly restrictive
or, at the very least, confusing patent claims. Excessive restriction and
confusion definitely does not foster innovation, as patent law was originally
intended to do.

However, from purely economic and motivational perspectives, individuals and
corporations still need concrete protection for truly inventive software.
Otherwise the incentive to innovate evaporates. We need to protect the rewards
accruing from innovation without hampering further innovation by denying or
restricting fair access to existing technology.

To achieve this, I believe patent law needs an overhaul, either to introduce a
new class of software patent, or to tighten the provisions of existing law, or
some combination of both. Here are some changes I would propose:

  • Reduce a software patent's effective period to five (or 10 at most) years.
    With the pace of change in the software world, a 20-year patent may as well be
    forever.

Require very specific definition of the "thing" being patented, especially
if the "thing" is a software invention. Patenting general concepts such as
"database," "operating system" or "hyperlink" is patent (again, I couldn't
resist) nonsense.Lapse patents that are not actively pursued. This "use it or lose it"
approach would prevent speculative patenting and discourage businesses that
specialize in the accumulation of potentially lucrative patents.

Make patent licensing subject to monopoly regulations, (such that a
company that refuses to license a patented technology, or that asks for
unreasonable license terms, could be subject to monopoly investigation).

I don't pretend that these simple suggestions would be easy to frame into
workable legislation, nor do I claim that they would completely eliminate
patent abuse, or address every flaw in the current system.

However, of this much I am sure: The current system is flawed, and
those flaws are in urgent need of redress.

John Fitzgibbon is a software developer from San Francisco, California.

"Commentary" articles are contributed by Linux.com and NewsForge.com readers. The opinions they contain are strictly those held by their authors, and may not be the same as those held by OSDN management. We welcome "Commentary" contributions from anyone who deals with Linux and Open Source at any level, whether as a corporate officer; as a programmer or sysadmin; or as a home/office desktop user. If you would like to write one, please email editors@newsforge.com with "Commentary" in the subject line.

Click Here!