[ih] History of duplicate address tests

[Brian E Carpenter]

On 30-Nov-22 09:23, John Kristoff via Internet-history wrote:
> My earliest memory, because that is the time of my entry into the
> field, of duplicate address detection was done on Token Ring networks
> during ring insertion (phase 2 - address verification).  More recently
> IPv6 has duplicate address detection (dad).
> 
> I'm curious to know about the earliest of duplicate address detection
> mechanisms.  I imagine there were some predating what I remember about
> Token Ring, but this is the kind of thing that is difficult to uncover
> without some hints. Any pointers to the earliest of these functions in
> networking?

I found myself asking whether the Cambridge Ring had, or needed, DAD, but
then I asked myself: is this a layer 3 issue, or did it arise in layer 2
designs (pre-Ethernet)? It wasn't needed, in theory, by networks that
generated layer 3 addresses mechanically from layer 2 addresses. How did
the orginal Appletalk deal with this?

So I have more questions than answers, I'm afraid.

(The multicasts caused by IPv6 DAD are a severe scaling issue for
large or low-power wireless networks.)

Regards
     Brian