[aprssig] Modification to AE5PL's routing idea

Jason Winningham jdw at eng.uah.edu
Tue Feb 15 11:29:10 CST 2005

On Feb 15, 2005, at 10:54 AM, AE5PL Lists wrote:

> True, that is part of the point.  Right now, we are trying to tell 
> users
> to do this which is even worse.

Absolutely.  That's one of the underlying problems, relying on end 
users to configure correctly.

> In fact, a SAR operation could be
> improved because they would be their own "private" LAN and multiple
> digipeaters could be easily preconfigured.

I'd think it would be advantageous to use portable digis to augment an 
existing network; if you want a "private" LAN, QSY.  Not really 
pertinent to the discussion at hand, though.

> You can't do that today with any of the digipeaters.  You have to
> configure them all.  This is actually a simpler configuration be cause
> you are simply saying "digipeat all directly heards and all directly
> digipeated by X".

I didn't state that very well.  I didn't mean take them out of the box 
new and turn 'em on, I meant take them out of the "go-box" and turn 
them on.  A generic config, be it RELAY, hopcount or whatever, is not 
dependent on physical location.  Once you add the "directly digipeated 
by X" requirement, this is a configuration parameter that must be 
entered on the digi before it is operational in a particular geographic 

I'd think our hypothetical SAR team would not have time to mess with 
this, and would not want to give up any access to the existing APRS 

> No, not necessary.  At most we would be looking at maybe 3 or 4
> digipeaters on the "ok to digipeat" list.  In most cases, probably only
> zero, one, or two.  This is not difficult and in fact, very powerful.

This would make it work for an area supported by the specific digis 
listed.  I don't guess it's too likely to operate outside a particular 

I guess my unease with this method boils down to "less configuration is 
better than more configuration".  It seems that it could result in the 
most effective bandwidth utilization of any method I've seen proposed, 
but it also requires the most extensive digi configuration, and could 
result in easily isolated LAN segments if a particular digi faults.


More information about the aprssig mailing list