[ih] TTL [was Exterior Gateway Protocol]

[Louis Mamakos]

On 6 Sep 2020, at 2:40, Jack Haverty via Internet-history wrote:

> The early gateways, IIRC, used the terminology "time", but in practice
> used just hop counts, since time measurements were difficult to
> implement.   The exception to that may be Dave Mills' Fuzzballs, since
> Dave was the implementor most interested in time and making precise
> measurements of network behavior.   I *think* Dave may have used time
> values and delay-based routing amongst his "fuzzies".

The Fuzzballs used a delay based metric in their HELLO routing
protocol. These delays were computed by way of the exchange of the
routing protocols messages, using some timestamps and as was also
used to sync the clock on the Fuzzball's.

A very similar mechanism was used in NTP later on, with some of
the earlier work in HELLO carried forward.  Separate from the time
sync protocols (but closed related) was the virtual clock model
in the Fuzzball code that the clock synchronization acted on.

The HELLO protocol was a distance-vector protocol; I forget now
what "infinity" was; perhaps 30 seconds?  It's been too many
years since I looked at that code.  I don't believe that the
forwarding code in the Fuzzball did anything clever with the TTL
during a forwarding operation, other than the usual decrement by
one.

louie