ietf
[Top] [All Lists]

Re: NATs *ARE* evil!

2000-12-18 01:30:02

In message 
<5(_dot_)0(_dot_)0(_dot_)25(_dot_)2(_dot_)20001217142128(_dot_)00a4f2f0(_at_)gnat(_dot_)inet(_dot_)org>,
 RJ Atkinson type
d:

At 13:32 17/12/00, Perry E. Metzger wrote:

       From an operator perspective, supporting *2* IP protocols
is much harder than supporting just one.  If one looks around,
very few NOCs on the planet today could reasonably be called
"fully successful" just managing IPv4.

lets not forget that supporting IP at all is counter intuitive - the
business case for providing interconnectivity to _other_ peoples
services is not there, unless they are already - it requires socialist
(i.e. government subsidy) to make the internet what it is now - i am
betting that most telco-child ISPs love NATs and simialr technology
coz they promote walled gardens (like wap etc) and lock in and all
that old stuff, BUT the only way out of this deadend that is IP v4
requires either OneBigIsp to take the plunge as a way of getting more
check marks on their service, or as a way of getting an even bigger
wall aroudn their garden (e.g. 3G guys might consider this:-),
OR it requires some sort of socially responsible behaviour (e.g. most
the 6bone is probably subsidized) to glop it all together and just
make it inevitable....(this is not specially a v6 plug, just a plea
for connectivity)
 
       So, if one wanted IPv6 to be promoted by operators,
one might spend time listening to operators and devising
clever ways to make multi-homing and routing work visibly
*better* with IPv6, to compensate for the much increased
operational burden.  Oddly enough, some folk are doing just
this.

indeed - we might as well work with what we have - hey, there's a lot
of stuff one can do with the code still, including just re-writing a
lot of cruft in routing code - btw, the way 3G access networks work,
one ought to be able to mandate for really good aggregation - i think
a lot of people forget that the exponential growth curve of the internet 
is not made out of homogeneous pieces - its actually a series of
technology changes - the phases from government and unviersity and
dial up, and dsl, and cable modem, and now mobile are all subtly
different (as witness they have their own ISP cultures and own address
allocation and routing headaches) - while datagram is one size fits
all as a way of communicating, we need a range of address alloc and
routing techniques for the different and future access and core
networks....i thought most this was discussed in the whole ipng debate
and was why we solicited input so widely....so now lets go (finish)
coding it


 cheers

   jon



<Prev in Thread] Current Thread [Next in Thread>