hi,
the old UK reverse name notation actually comes from some sensible
ideas - firstly from the big-endian processing methods - but also the
most important part of the address
comes first - ideal for global routing decisions early. who cares
about the actual hostname , get to the actual TLD ;-)
anyway, a little unfair as that decision was made before the Internet
domain standard was agreed/established. hey, competing systems...one
of them usually wins. in this case the
other one did ;-)
as for IPv6, the topic of this thread. having done campus IPv6
deployments, working out addressing schemes, sorting out kit upgrades
(and broken by many 'oh, IPv6 is in a future
release' or 'its on our roadmap' vendor promises) a few things. it
gives us native end to end on a network that is now too big to handle
that with IPv4 - NAT etc causing all kinds of new
things to be cooked up to ensure things dont break. deploying it is
trivial-ish (these days) - you have so much choice...and eventually
decent routers doing SLAAC will finally be able to serve
other details such as DNS/time/etc via SLAAC - servers? give them
static addresses...simple ones that dont populate all the last half...
that gets me on to my small annoyance... /64 bit subnet masks for
local networks. really? ALL of that address space and then throw such
a large range away on subnets commonly populated
with no more than a couple of hundred clients...maybe a few thousand
at worst. what a mistake.
I come from a background where we had IPv4/DECNET/AppleTalk/IPX all
around the place - to be honest, 2 fairly simply IP protocols being
handled/routed has never kept me up at night
and I enjoyed many times of cleaning things up and getting people to
realise what access their systems needed...a quick refresh of access
rules (on hosts and in network kit) and
monitoring ('you monitor that service on its IPV4...why not IPv6' was
said way too many times)
address format? at least you can put :c01d:c0ff:ee and dead:beef etc
in your addresses... as others have said, IPv4 is only a number in a
superficial sense (who HASNT been burnt
by an engineer putting a few 0's into IP address boxes on kit that
forces all fields to be populated? we had A6 and AAAA mess, things
took a while to iron out and just like
BSD dying, IPv6 deployment (and DNSSEC!) just really hasnt been
'completed' yet. but thats okay, because I'm still curious why the US
techies didnt just bite the bullet and
got for IPv8 ;-)
alan
