[79594] in North American Network Operators' Group
RE: djbdns: An alternative to BIND
daemon@ATHENA.MIT.EDU (andrew2@one.net)
Mon Apr 11 17:26:53 2005
From: <andrew2@one.net>
To: <nanog@merit.edu>
Date: Mon, 11 Apr 2005 17:26:29 -0400
In-Reply-To: <Pine.LNX.4.44.0504111657480.14707-100000@localhost.localdomain>
Errors-To: owner-nanog@merit.edu
owner-nanog@merit.edu wrote:
>> however, since BIND9 is compatible with BIND8 and BIND4, and with
>> microsoft's DNS, and with virtually every other DNS in the world
>> except for "tinydns",
>
> Err, "compatible" because it detects them and then does the
> right thing, and uses the traditional protocol.
You know...I'm reminded of something we're all familiar with that came
up, oh...lets say 8 years ago. There were some new-fangled devices out
there that were capable of communicating over POTS at somewhere close to
56 kbps. It seems to me there were two flavors of them, K-Flex and X2.
You might have heard of them. Anyway, if your modem had K-Flex firmware
and was trying to connect to something using X2, you couldn't connect
anywhere near 56 kbps. And vice-versa. The two technologies were
incompatible. And yet, once they detected the incompatability, they
were able to renegotiate down to a protocol they had in common, say
v.32. Now eventually we came out with the v.90 standard so that
everyone could play together nicely. Point is, even before there *was*
a 56k standard, all those "incompatible" modems could still communicate,
just not using their new proprietary protocols. So, I guess I'm
wondering....how is what BIND9 does substantially different than the
case I've outlined above?
Andrew
