Turkey has switched Root-Servers

Stephen Sprunk stephen at sprunk.org
Wed Sep 28 02:07:02 UTC 2005


Thus spake "Peter Dambier" <peter at peter-dambier.de>
> Christopher L. Morrow wrote:
> > I'm confused by the reasoning behind this public-root (alternate root)
> > problem... It seems to me (minus crazy-pills of course) that there is no
> > way for it to work, ever. So why keep trying to push it and break other
> > things along the way?
>
> Paul Vixie has given very good arguments.
>
> Let me add a design fault:
>
> As more than 80% of all names are registered under '.com' there is no need
> for any other domain.
>
> Ok, let us get rid of all those domains and put them under '.com.
>
> Now there is no more need for '.com' either. Let us get rid of it and
> we have finally got more than 3000 toplevel domains. That is all we want.

No, what you'd get is 25M top-level domains and virtually no hierarchy.
That is _not_ what we want.

.com is an abomination, as are the other gTLDs to a lesser extent.  .gov,
.mil, .edu, .info, and .biz need to be shifted under .us immediately, and
everyone under .com, .net, and .org needs to be gradually moved under the
appropriate part of the real DNS tree.  I can live with .int continuing on,
but only because no better solution immediately comes to mind.

> Let me compare Public-Root and ICANNs root:
...
> The Public-Root has got 3043 domains. ICANNs root has got only 263.
>
> There is a political design problem with ICANNs root. It has not got
> enough toplevel domains.
>
> DNS was designed as a tree. It was designed decentralised.
>
> DNS today has degenerated to a flat file like /etc/hosts was.

What you're proposing is eliminating what little tree-like elements are left
and making a totally flat system.  Can't you see that you're arguing against
your own position here?

S

Stephen Sprunk        "Stupid people surround themselves with smart
CCIE #3723           people.  Smart people surround themselves with
K5SSS         smart people who disagree with them."  --Aaron Sorkin




More information about the NANOG mailing list