Lively discussion

[David Hambly]

Folks,

Glad to see all this discussion.  What made sense to me so far was a Linux 
Winlink client (at the network layer Winlink is an open system) but don't 
stop there.

IMHO the digital mode question should look like a swiss army knife.  The 
comm layer should be flexible, allowing for Packet, PacTOR l, MT63 ARQ 
(granted vaporware), maybe FS-1052 ARQ as soundcard modes evolve, and they 
will, this modularity would support them as well as legacy systems.

What is also important is network establishment (NE), maybe something as 
simple as FNpsk or as complicated as ALE.

Checking in, listing traffic, accepting traffic, checking out and delivering 
traffic is what nets do.  This traffic handling framework will likely not 
change much.

Having a Linux Winlink client should be a part of this, seperate the data 
(digital mode) engine and the network establishment protocol and you are 
still left with the question:  What to do/has been done with this traffic?  
Traffic comes from many sources and, in the event of certain comms failures, 
arrives by different means.   It's not an all digital world, some traffic 
will come by voice and NetManager allows for manual entries; but what about 
automagic entries...from different digital modes/sources?  The ability to 
account for traffic however it comes and wherever it goes.

So I'm thinking in terms of digital mode, network establishment AND traffic 
accounting.  ALE does a fine job of the first two but I think the idea of 
integrating traffic accounting completes the solution.

If anyone actually takes this on...watch those dependancies, take a look at 
the XARPM installation, worked great on my old RedHat 6.0...no 
searching/upgrading kernels/libs/window managers.  Why is this important?  
Because Linux can turn an old PC into a emcomm warrior, this keeps the cost 
low.

73
Dave - KG6QG