Horace Heffner wrote:
I can't image any decision that could be more damaging to the US
software industry, especially in the global marketplace.
NO, Horace, this is GOOD NEWS!! GREAT news!
Software patents are horrible, and choke innovation. They do *not* help
innovation in any way shape or form.
It's normal for the guts of innovative for-profit software to be kept
proprietary, and the code can be copyrighted. This has *not* changed.
Twenty five or thirty years ago, when software innovation was galloping
along (spreadsheets had just been invented, the Xerox Star had sparked
the desktop revolution, the Mac was just coming out), there WERE NO
SOFTWARE PATENTS. Software was considered to be an "algorithm" and as
such was mathematical in its character, and was no more "patentable"
than the quadratic formula.
Lotus 1-2-3, and, later, Excel, would have been impossible had there
been modern software patents: VisiCalc's interface would have been
patented and Personal Software would have sued the pants off Mitch Kapor
and shut down the whole field, and we'd all still be using VisiCalc.
The main innovation 1-2-3 had was speed, and that would not have gotten
them out from under an electronic spreadsheet patent. But there was no
such patent: VisiCalc was copyrighted but not patented.
Then shortly after that, the "look and feel" patents were accepted, and
huge amounts of time and effort started going down the rathole of patent
fights, in particular between Microsoft and Apple. Did this improve or
increase innovation on the desktop? Not that I can see!! Quite the
contrary; things have been pretty frozen ever since in the interfaces
world, which is where those patents first "bit". Oh, there have been
minor tweaks, and the underlying software has gotten better
incrementally, but the overall "look" is about the same as it was when
it first became possible to patent it. However, there have, in fact,
been really substantial changes in the "open source" look and feel,
where nothing's patented -- the look of an X windows desktop today is
very very different from how it looked a decade or so back. I'm not
sure one can say the same for a Windows desktop -- that still seems
pretty similar, in overall outline, to the way it looked when Windows 95
was new.
And then we got the SCO effect. SCO claimed to have a patent on the
core technology of Unix, and started suing companies to make them pay up
(SCO was by that time a company consisting mostly of lawyers, producing
nothing but lawsuits and profits -- zero innovation from _them_!). IBM
and Microsoft accumulated arsenals of software patents; there was a cold
war going on between them in which there was a threat of global meltdown
if they started enforcing their outrageous patents on everything from
page faults to sorting algorithms.
GIFs were patented; were you aware of that? Rather, the LZW compression
algorithm was patented. It resulted in a lot of awkwardness and wasted
time among people who wanted to avoid infringing the patent. Did any
developers *benefit* from this patent? No, not at all -- it was owned
by some large company (SCO again, maybe? not sure) which was not in the
business of innovating; they were in the business of litigating.
The attempt to patent the page fault algorithm some time back was the
most horrendous of the patent problems; I'm not sure how that shook out.
In general, proving "prior art" for software algorithms seems to be
difficult; certainly a lot of the garbage that's been patented was
already developed elsewhere, or was developed near-simultaneously by
multiple individuals at different companies, all basing their work on
the same "ambient" knowledge base. But challenging a patent on those
grounds doesn't seem to work very well, if I can judge by what I could
see of the results. The LZW algorithm, and the consequent GIF patent,
are a case in point: LZW is a minor variation on a common idea in data
compression. Certainly, I recall conversations about the approach of
using back-references to do compression from a time long before the LZW
algorithm was published. The difference is Lempel and Ziv actually
published it, and then they -- or more likely the lab where they worked
-- applied for a patent; other folks working with what amounted to the
same algorithm didn't think to do that. And then the patent ended up in
the hands of a bunch of lawyers who treated it as a cash cow. Just how
did that encourage innovation? In a word: It didn't.
This is
something that demands immediate congressional attention:
God I hope not!
http://www.patentlyo.com/patent/2008/07/the-death-of-go.html
The loss of the method patent amounts to the loss of thousands of jobs,
of an entire industry.
What, they'll have to fire all the lawyers?
Are we really so scared that Google will get some competition?
This statement:
"It will surely be cause for mourning among those who believe that
allowing patents on cutting edge technologies has served the country
well for more than two centuries"
is B.S. The software patent is a new concept and has benefited a few
large players, like Google, and a whole lot of patent lawyers.
For centuries we had the notion that you can't patent an algorithm, and
that served us well. The *change* was the inversion of that idea, to
allow big companies to lock out competitors by use of algorithm patents.
It turns vast amounts of case law on its head.
This is huge, and it could impact cold fusion protection as well, when
it finally reaches the stage of practicality.
Best regards,
Horace Heffner
http://www.mtaonline.net/~hheffner/
