From: Henning Schulzrinne <schulzrinne(_at_)cs(_dot_)columbia(_dot_)edu>
To combine the two long-running threads: The solution to the NAT problem
is obvious - we need a submarine patent where somebody claims rights to
NATs and then charges so much for licensing that it makes technically
more sound solutions, say, IPv6, economically attractive. Indeed, I
think we should get together a group of people to patent all the
architecturally bad ideas (call it the "RSI group"), as they'll appear
sooner or later. That will give us 20 years of respite...
Beware of what you wish for.
It's far more likely that there is a so not widely unknown patent
on IPv6, such as the idea of using more than 32 bits in an address.
If you can get a patent much more obvious ideas, such as
http://patent.womplex.ibm.com/details?&pn=US05446889__ or
http://www.patents.ibm.com/details?&pn=US06025810__&s_all=1#23 then
the far more obscure and (possibly?) more practical idea of a bigger
transport layer address is surely patentable.
For example, remember the Soderblum Patent that caused so much
grief for FDDI. (see http://www.patents.ibm.com/details?&pn10=US04293948 )
Vernon Schryver vjs(_at_)rhyolite(_dot_)com