[aprssig] Definitive digipeater specification?

[Tapio Sokura]

scott at opentrac.org wrote:
> Is there any 'official' explanation of exactly how a modern APRS digipeater is supposed to function?  The AX.25 spec just covers the conventional digipeating.  I think I have a pretty good handle on it, but I know there's been some inconsistency out there.  Just want to make sure I'm working from the best source material available.

One thing missing from most, if not all, current digipeater 
implementations is the general ability to dynamically adjust to the 
amount of traffic and band propagation conditions. I haven't made any 
practical experiments or even theoretical calculations on this, but at 
least it shouldn't hurt if the digi could limit digipeating far away 
stations when there is a lot of local traffic, for example.

I'm not talking about just counting the digipeater hops, but also taking 
into account the actual geographic distance to the originating station, 
calculated from knowing the digipeater's location and parsing the 
location of the remote station from of the candidate packet. There are a 
lot of other things a really smart digipeater could do, if it peeked 
further into the packet contents after the AX.25 header. The KPC-3 can't 
do this, but it shouldn't stop more powerful digipeaters from developing 
further from where KPC-3 left off (= the previous millennium).

   Tapio