Date: Wed, 18 Dec 1996 23:35:58 -0800 (PST)
From: "Gregory J. Woodhouse" <gjw@wnetc.com>
Sender: gjw@shellx.best.com
To: Keith Moore <moore@cs.utk.edu>
cc: cos@leftbank.com, drums@cs.utk.edu
Subject: Re: The string after HELO 
In-Reply-To: <199612190712.CAA26883@ig.cs.utk.edu>
Message-ID: <Pine.SGI.3.95.961218232237.8219O-100000@shellx.best.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII

On Thu, 19 Dec 1996, Keith Moore wrote:

> 
> I would think that a stable name of any kind that identified the
> actual host would be better than either a temporary IP address
> or a temporary domain name.  (but how does a host know whether
> its domain name is temporary?)
> 
> Keith
> 

I agree that a stable name is preferable to a dynamically assigned IP
address (and potentially an associated domain name). In response to your
question, a host does know whether or not it receives its IP address though
dynamic assignment (as with DHCP). In such a situation, I'd simply presume
that the domain name is not permanent. Now, as it happens, where I work we
use DHCP to assign IP addresses to workstations (servers have fixed IP
addresses), but when my lease expires on my IP address it is always renewed
with the same IP address, so in practice I get lazy and treat it a if
though it were static. But, then again, I know it could essentially change
at any time.

In a sense the question is moot because mesage routing is not possible in a
situation such as this--unless there is  second domain service that assigns
stable names to workstation. (Mail could go to a  central server, but then
there's not much point in using SMTP for final delivery when better methods
such as IMAP exist for message access.)  

---
gjw@wnetc.com    /    http://www.wnetc.com/home.html
If you're going to reinvent the wheel, at least try to come
to come up with a better one.