In message <[EMAIL PROTECTED]>, Sean Doran typed:
>>Wait, it's because of *me* that IPv6 isn't a stunning success compared to NAT?
>
>>I didn't realize that, when I asked the IAB to use their technical insights
>>as a market predictor, that behind the invisible hand of the marketplace
>>lurks little ol' me. Maybe someone should tell Paul Krugman.
Pagan God of Market Forces...
oh thats no good - i want to speak to the 21st century god of market
forces....and of course, she retired as uk primeminsiter a while
back...
so NATs address a user requirement as WELL as a network provider
requirement - for example, asymwetric nats let the user get away with
sloppier security (not just think they can ) by reducing the available
services
of course you could argue the same for ipv6 given your view that
noones useing it in that unlike an asymetric nat (i can reach you but
you can't reac me) v6 is symettreic in its ability to disable people
:-)
but fantasy aside, ipv6 involves _routing_ as well as addressing - NAT
reduces the problem space for providers and users by reducing the
number of services reachable - so? thats not exactly comparing things
on a level playing field is it?
to make v6 work tarks end users more work than v4 (they have to
(re(re)-number) and worry about security policies properly) and takes
router vendors some work to write some working code (something they
have few people left able to do - sure everyone else has fewer
still)).
so it isnt surprising that one has seen more deploym,ent than the
other
, but it is not a measure of the releative (de)merits
of NATs and ipv6
thats just in yr. head....
cheers
jon
i'd like to think we could talk about technical things we CAN do to
maintain and increase connectivity, as well as retaining or decreasing
the ease of config for an end user - to some extent, its one of those
threshold things - despite its shortcomings, if people can get over
the v6 threshold, it might sort things out provided a half way decent
addressing plan AND retaining NAT type functionality for various
reasons.....of course there's multihomeing to worry about, but nothing
in any other scheme 've seen sorts this - its an inherently hard
problem to provide "always on" addresses, aggregation/hierarchy for scaling,
and multihomeing - some sort of set theory #101 makes this obvious...
j.
p.p.s
why has noone addressed my replace the DNS comment ? :-)
