Re: Motivation to submit an idea in IETF?2010-01-22 01:59:49On Jan 21, 2010, at 7:51 PM, Greg Daley wrote: I think that whetever the reason, documents submitted to the IETF are less likely to become standards track RFCs if there is critical IPR which must be licensed in order to construct the protocol. As somebody who makes a living explaining patents to people who feel threatened by an infringement suit, the simple fact is that the IETF is appallingly (and blissfully) ignorant of the vast number of patents that are filed relating to our RFCs. In general, we do not discover that there is critical IPR until well after the protocol is published, then we get mad about it. One rational for publishing rather than patenting is cost. If the goal is to assure your rights to use your own invention rather than paying a fee to a patent troll for that use, thorough publication of the invention as early as possible may be a cost-effective solution in countries whose regimes discredit 3rd-party patents of already- published ideas. I've used this technique fairly effectively in patent- busting in the US. I've even used mailing-list records to provide invalidation arguments. Writing an internet-draft describing the invention is cheap. Filing a patent application is expensive. There are documentation fees, legal fees, artwork fees, filing fees, maintenance fees, patent search fees, and the ongoing hassle and expense of examination, re-examination, forced continuations, and so on. You might spend upwards of $50,000 to get a patent (although some do cost much less, especially the do-it- yourself filings). And then you have to enforce the patent, and that's not cheap; you have to look for people who might voluntarily license it and negotiate with them, and you have to look for people who have infringed it and work through a more adversarial negotiation or even an infringement suit. I've seen infringement suits run above half a million dollars. But if you have a patent, don't maintain it, and don't enforce it, it gets weaker or might even completely evaporate. Sure, it serves as strong prior art to keep some other bozo from patenting the exact same thing even if you let the patent "go to seed", but all they have to do is tack on one important additional claim that your didn't foresee, and you're back to having to pay the troll. A submitted application is strong prior art. A published paper or internet-draft is pretty good, but not as good. A really clear write- up in mailing-list email can be helpful, especially if the "inventor of record" (the person whose name is on the patent app) is an active participant on that mailing list. So it all comes down to intent: If your business is a real participant in the industry and you're going to make your money off of building and selling a product that includes the invention, you may get adequate protection for your business by publishing the invention early on. If you're a big fish in the industry, then you probably can benefit from having a pool of patents for cross-licensing with the other big fish, so filing might be useful. And of course, if you're not really participating in the industry but hope to make money off of licensing the invention, then filing is the only way to go, although it's harder to make money off such a patent than you might think. In: "A Farewell to Alms: A Brief Economic History of the World." (Princeton, NJ: Princeton University Press, 2007. xii + 420 pp. $30 (cloth), ISBN: 978-0-691-12135-2), the author Gregory Clark cites a number of cases relating to patented inventions in the textile industry, where the inventors who relied on patents for their funding died penniless, but the people who just built things and made product with them experienced economic success. It's a good lesson to us all. Its' a "good read" as well, and I highly recommend it; my depressing theory, however, is that we're falling off the tail of the success hump and sliding back into a strictly Malthusian model of supply, demand, and starvation. -- Dean _______________________________________________ Ietf mailing list Ietf(_at_)ietf(_dot_)org https://www.ietf.org/mailman/listinfo/ietf
|
|
||||||||||||||||