RE: regexless parsing, again?

Eric Fitzgerald wrote:
>The terms "obvious" and "prior art" come to mind.

Everyone knows that the patent system is horribly awry with
regards software. Like many of the old-timers who started
off with the security industry, I get 2 or 3 patent-related
enquiries (for or against) each year - it's ridiculous because
most of the patents are bogus. I've been involved in 5
cases in the last 7 years regarding patents granted
for intrusion detection systems. PTO keeps issuing
patents for rocket science ideas like using statistics
on packets (!wow!) and in-line blocking (*gasp*!) of
hostile traffic. The field of logging is just as bad, except
that the prior art here stretches back to the 1970's
where as IDS didn't have a history prior to the mid 1980's.

None of that changes anything. Several of the lawyers
I've worked with on patent issues have been quite
cheerful about the broken-ness of the patent
system. After all, they litigate patents for a living,
and if the system weren't broken, they'd have to go
chase ambulances or sniff asbestos or something
else. The PTO is judged by how many patent
applications it examines and how many patents
it grants, so it's actually got an incentive to keep
issuing bad patents. Especially since most PTO
lawyers know they can get a job at a patent
specialist law firm after a few years doing the
government grind. The entire system is in a
huge feedback loop that guarantees the problem
will get worse and worse.

Meanwhile, for a normal person - the proverbial
"little guy" it doesn't matter if the patents are
bogus or not, because the cost of defending
yourself if you're sued by a company with one
of these bogus patents is now prohibitively high.
And, like with the asbestos litigation, there are
now jurisdictions that specialize in being a
friendly place for patent litigation. If you want
to defend yourself against a patent lawsuit from
a big company with a bogus patent, you'll spend
at least $500,000 or more in the first year. And
there are whole strategies - the holders of these
bogus patents try to get a couple of start-ups
to settle (establishing some kind of precedent)
before they approach Cisco or IBM. Try to
squeeze enough money from the little guys to
be able to afford to tackle the big boys. And
meanwhile you've got experts like me pointing
out shelf-space of prior art - but that has
absolutely no bearing on anything. It's more a
question of who has the financial fuel and the
better legal strategy.

But the patent system is not completely wasted.
Last winter I ran out of fuel for my outdoor wood-burning
furnace, and burned 6 boxes of patent documents in
it. Kept my house toasty for 3 whole days, it did.
I shudder to think how much that fuel cost somebody,
though...

mjr.