Paul G. Allen wrote:
The only reason we still have such a problem with software patents is because the courts have no concept of it, and the lawyers are blood suckers and will go where the money is.
I think there's more to it than that. Certainly the patent examiners have a decent grasp of the technology and argue that some aspects aren't novel.
I think part of the problem is that a software patent tends to be too "soft."
If you patent a new windshield wiper design, there's no way that can be interpreted to be a patent on a novel cancer treatment. However, I've seen all kinds of arguments along the lines that (say) a network routing algorithm patent covers an electronic commerce application, because you can interpret the MAC address as a user name, and an SSL master secret as a user authentication password, and the merchant might actually be an intermediate router, and the card number might be represented as an IP datagram, and stupid crap like that. When the software patent starts getting read like a patent on algebra, with arbitrary substitutions of items for other items regardless of how nonsensical it is, you wind up having software patents being problematic. I've even seen "software patents" used to try to block hardware development, using the same techniques of redefining elements of the claims to apply to completely nonsensical but isomorphic bits of the hardware.
What kind of software patent isn't problematic? I think stuff like the DH and RSA patents are excellent examples. I think patents on particular data structures might make sense. Not that I necessarily approve - of course I'd rather have the research free for the taking, but I can see where patenting a particular algorithm for routing, cryptography, synchronization, or something like that could make sense, if it doesn't get applied to completely irrelevant fields.
Even hardware patents suck. For example, the patent on the segway covers any vehicle with two wheels on the same axle facing perpendicular to the axle, with the center of gravity above the axle. No discussion of how to balance it, power it, steer it, or anything else. There's nothing in the patent that you couldn't learn by watching a 10-second video of someone riding a segway. But, and here's my point, you can't possibly interpret it to apply to an airplane, a house, or a farming technique.
-- Darren New / San Diego, CA, USA (PST) -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
