Bob Braden <braden(_at_)ISI(_dot_)EDU> writes:
*>
*> So, on the one hand, we have the actual behavior of millions of people.
*> On the other hand we have Keith Moore's opinion about what they ought
*> to prefer. I don't have any trouble figuring out which one I believe.
*>
*> -Ekr
*>
Erik,
Eric.
Errr, let's see if I understand your argument. You advocate boxes that
have the effect of disallowing a large class of services. We all
understand that such boxes are in fact in heavy use in today's
Internet, so there is a large set of users who cannot use any of the
services from this class. So, your argument is that you can
demonstrate that users don't want any of these blocked services,
because they aren't using them today.
Huh?
Repeating this argument endlessly and fervently does not make
it any less a contradiction. Can we move on to something more useful?
Instead, maybe we should start by you taking the time to understand
the argument properly.
Users aren't physically handcuffed to their Internet connections.
They have choices as to who to purchase connectivity from. Those
users, if they chose, could purchase connectivity with static IP
addresses and no NAT. They by and large don't do so. Therefore, it's
reasonable to conclude that they don't in fact want them. Or, to put
it more precisely, that the marginal disutility of getting rid of NAT
exceeds the marginal utility of these new services.
Now, the situation is a little different when we're talking about
people who are on corporate networks, but I view the corporation,
not the users, as the customer here and so the same logic applies.
If the corporation wished the users to have those services enough
to disable NAT, then they would do so.
-Ekr
P.S. And btw, I'm not advocating NAT. What I'm advocating is that
we stop behaving as if we think that anyone who uses NAT is obviously
an idiot.
--
[Eric Rescorla ekr(_at_)rtfm(_dot_)com]
Web Log: http://www.rtfm.com/movabletype