Re: DS MIB - please review this list...

[Brian E Carpenter <brian@hursley.ibm.com>]

Hmm. My computer science instinct says that having them adjacent doesn't 
do a thing for generic usage - you can't rely on it - it would need to be
a single nested data structure. But if this is the conventional wisdom
I suppose we have to live with it.

  Brian

Bill Fenner wrote:
> 
> >Can you argue the case for the stricture, generally or (better) in this
> >application?
> 
> Fred,
> 
>   Here's the answer I got when I asked a similar question.
> 
>   Bill
> 
> ----- Begin forwarded message:
> 
> From: Juergen Schoenwaelder <schoenw@ibr.cs.tu-bs.de>
> Subject: Re: draft-ops-endpoint-mib-04.txt
> Date: Wed, 12 Jan 2000 11:45:53 +0100
> To: fenner@research.att.com
> Cc: daniele@zk3.dec.com, mibs@ops.ietf.org
> 
> >>>>> Bill Fenner writes:
> 
> Bill> A related question: if there are multiple addresses in a given
> Bill> table row, where all of these addresses must be the same type
> Bill> because of the way the thing being managed works, do you have to
> Bill> have an InetAddressType for each?
> 
> I would say yes. And I would even strongly suggest that the
> InetAddressType and the InetAddress columns sit next to each other in
> the order (InetAddressType, InetAddress). This rule will ensure that
> generic managers will be able to understand the relationship between
> these objects (which we unfortunately can't represent formally in the
> SMIv2).
> 
> /js
> 
> --
> Juergen Schoenwaelder      Technical University Braunschweig
> <schoenw@ibr.cs.tu-bs.de>  Dept. Operating Systems & Computer Networks
> Phone: +49 531 391 3289    Bueltenweg 74/75, 38106 Braunschweig, Germany
> Fax:   +49 531 391 5936    <URL:http://www.ibr.cs.tu-bs.de/~schoenw/>
> 
> ----- End forwarded message: