Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!orion.oac.uci.edu!ucivax!gateway
From: hta@isolde.er.sintef.no (Harald Tveit Alvestrand)
Newsgroups: comp.protocols.iso.x400
Subject: Re: Implementing generic service in the Internet
Message-ID: <1991May27.074707.6004@ugle.unit.no>
Date: 27 May 91 22:07:00 GMT
References: <1991May24.013826.7713@mel.dit.csiro.au>
Reply-To: harald.alvestrand@delab.sintef.no
Organization: ELAB-RUNIT, SINTEF, Norway
Lines: 18
Approved: usenet@ics.uci.edu
x-attn: jns
ReSent-From: Re-sent but not originated by Jerry Sweet <mhsnews-request@ics.uci.edu>
ReSent-To: mhsnews@ics.uci.edu

In article <1991May24.013826.7713@mel.dit.csiro.au>, smart@manta.mel.dit.csiro.au (Robert Smart) writes up an idea for a new way to handle "generic services"
by making a specific IP subnet the "place to be" for generic services, and
adding a "discovery protocol".

Question:
How is this different from adding multiple service records to the DNS
(in the mail instance, multiple MX records for x400.net) and making
the users of this calculate the IP distance?

Yes, I know......stupid machines will go on using the first one.
But still, guaranteeing IP-address collisions and using the routing protocols
to carry service information seems rather a strange path to me.


                   Harald Tveit Alvestrand
Harald.Alvestrand@delab.sintef.no
C=no;PRMD=uninett;O=sintef;OU=delab;S=alvestrand;G=harald
+47 7 59 70 94