Re: [mdnsext] mDNSext features/requirements rollup

[Michael Richardson <[e-mail]>]

>>>>> "RJ" == RJ Atkinson <[e-mail]> writes:
    >> The problem I see is that, because of historical practice,
    >> attempting to treat (2) as a special case of (1) will mean that
    >> names designed to be local to (1) will leak into (3).

    RJ> The claim above is not obviously true for several reasons:

    RJ> A) Split-horizon DNS deployments have been around since the
    RJ> 1990s at least.  THese are widely deployed.  Few problems have
    RJ> arisen in practice, as a percentage of split-horizon
    RJ> deployments.  Yes, problems could arise, for example if a human
    RJ> misconfigures something, but it isn't inherent in a
    RJ> split-horizon DNS deployment.

There are significant problems with split-horizon.
They didn't show up until we had nomadic devices like laptops and now
smartphones.   Consider that the MIF WG has an entire draft to
attempting to solve this problem (I think it's a total fail though)

Big organizations have fewer problems because they know how to deploy
the right infrastructure and do the right configuration and training for
end(remote) notes, but smaller (dis)organizations basically devolve to
using IP address literals for all internal hosts ("intranet"s) because
of road warriors and VPNs and the like. 

That's why mDNS has been such a success, and why in homenet and here
there has been such a big discussion about what is .site, because nobody
wants to type IPv6 literals, and yet v6 global end addressability
is making people realize that they will want (some) connectivity.


-- 
Michael Richardson <[e-mail]>, Sandelman Software Works