[This is probably better off in PAGAN, as it isn't really specific to ARIN]
>Are other people coming up with more conservative (i.e. exhaustion occurs
>sooner) estimates? If so, and if exhaustion is projected by the millenium
>or thereabouts, then my premise would be incorrect.
First define "exhaustion". There are people who claim that we have
already exhausted the IPv4 Internet address space as there are viable
Internet technology using applications which cannot be implemented on
today's Internet due to address space limitiations (or so I'm told).
I tend to argue that Internet address space will NEVER be exhausted
for the simple reason that people depend on the Internet and as such,
they'll do whatever it takes to get the resources they _need_. The
persistent (as yet unsubstantiated) rumors of a black market in /16s
is merely one facet of people doing whatever it takes to meet their
More specifically, as address space becomes more and more scarce (or
rather, perceived to be more and more scarce), the "value" of that
address space will increase. As value increases, organizations
currently holding vast tracts of unused Internet address space will
have increased incentive to convert that address space into used, most
likely by someone else who would be willing to pay real money to
obtain the space.
Of course, such "non-traditional" approaches toward obtaining address
space are not recoginized under current registry policies.