[aprssig] "Superdigi" experiment at MB7UBN

[Kai Gunter Brandt]

Rick Green skrev:
> On Sun, 15 Mar 2009, Chris Moulding wrote:
> 
>> I'm currently running an experiment to test a "superdigi" option using 
>> the MB7UBN igate. The superdigi option digipeats WIDE1-1 as normal but 
>> also digipeats any other valid APRS packet and re-transmits it as a 
>> WIDE2-1 packet to allow it to be digipeated once more by another 
>> digipeater.
> 
> Somehow, I cringe at the thought of a digipeater modifying the path of an 
> incoming packet, especially in the direction of extending it, even one 
> more hop.  Not only will it increase channel congestion, but it also hides 
> obsolete or mis-configured TNCs, removing the incentive for an operator to 
> learn current best practices.
> 
> DANGER: You're on a slippery slope here.

I'll agree to this.

Regarding the ISS digipeating you should only gate packets from ISS and 
not local stations.

And when users usualy use a path of ARISS,SGATE,WIDEn-N or ARISS,WIDEn-N 
it should only forward a packet digipeated by iss.
So then a regular WIDEn-N and/or a digi on SGATE solves the problem.

ARISS,SGATE,WIDEn-N is going to be RS0ISS-4,SGATE,WIDEn-N when it's 
heard from ISS

Retransmitt all other valid packets are undermining the consept with a 
standard.

I can agree to temporarely allow "unwanted" paths with ONE jump but this 
could be done with UIDIGI.

> That said, I'm of the opinion that APRS messaging as currently implemented 
> is a horribly inefficient protocol, and I would welcome a digi option that 
> will simply refuse to retransmit any message packet that has already been 
> digi'd ONCE, no matter what path the originator is requesting.
> 

Its probably better to use time on a better message protocol.
Functions like UIDIGI, UIFLOOD and UITRACE are what we expect to be able 
to configure and are probably enough.

As there are messages that should be ACKed and some not it's probably a 
nice idea to see if it's possible to make a system handling duplicate 
messages.

Other functions could be similar to the uSmartDigi like digipeating only 
stations within a circular range or say it should not digipeat anything 
within a sector. (could be useful where several digis hear each other)

So simplicity rewls.. The TNC-X with XDigi has all the expected 
functions and it could be configured remote via radio..

So maybe more functions like that??

Kai Gunter
LA3QMA