[35395] in North American Network Operators' Group
Re: Namespaces (was: new.net: yet another dns namespace overlay
daemon@ATHENA.MIT.EDU (Vadim Antonov)
Thu Mar 8 06:25:44 2001
Date: Thu, 8 Mar 2001 03:23:23 -0800 (PST)
From: Vadim Antonov <avg@kotovnik.com>
To: Sean Donelan <sean@donelan.com>
Cc: nanog@merit.edu
In-Reply-To: <20010308102822.29348.cpmta@c004.sfo.cp.net>
Message-ID: <Pine.LNX.4.04.10103080318080.12592-100000@kitty.kotovnik.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Errors-To: owner-nanog-outgoing@merit.edu
On 8 Mar 2001, Sean Donelan wrote:
> Somewhere along the process, DNS changed from an address space to to subject
> space. As an address space, having globally-unique identifiers is important;
> but as a subject space, searching is more complicated because identifiers
> aren't unique.
That was inevitable. After all, FQDNs are human-readable.
"For every problem there is a simple and obvious solution. Usually that
solution is also wrong". Hierarchial name spaces is a choice example of
obvious and wrong solution to global naming.
> DNS is not, nor ever was intended to be a general purpose search tool. That
> was X.400/X.500's job :-)
Too bad they !*@!d it...
> If someone wanted to do something interesting, they would come up with a
> new RESOLVER library and interface which searched on something at a higher
> level than DNS names.
It is already done :) Yahoo, Google, Altavista, etc etc etc :)
Just stop issuing alhpanumeric domain names and use numerals only.
--vadim
