[aprssig] Weather Stations and Net Neutrality

[Robert Bruninga]

> The RF network is a very limited resource.
> ... when you have the opportunity to save some 
> bandwidth, at the cost of only a little extra  
> programming, it is worth it. That was the 
> opinion of everyone (besides Bob) ... when it 
> was proposed...

It was never worth it.  Positionless weather saves one packet
per hour but at the expense of unreliable delivery, and delays
and latency.  Miss one weatherless-postion packet and the next
almost two hours and a dozen positionless-weather packets are
useless to the newcomer, the mobile, the person who just
activated for a storm, etc.  An abomination in my mind to a real
time system!

And the one weatherless-position packet is hugely verbose too.
SO add up the bytes of the 7 packets per hour for the sum of a
positionless-weather station and the long verbose
weatherless-position packet.  Add up the total bytes
transmitted, add the fact that none of the positionless WX
packets are of any value if the weatherless-Position has not
been received, and that the weatherless-Position packet is
useless until a positionless-weather packet is received.  This
process loses by every measure of value to the APRS concept in a
collision based network.

On the other hand, the combined WX/POSITION packets provide the
highest reliability period, no matter how many may get lost or
collided, you only need to get one and you are up-to-date.  In
clude the original decay algorithm so that redundant benign
weather is transmitted less often, but new changing weather is
transmitted immediately and more often and you can see how well
APRS WX was supposed to work...  All of this was lost when the
positionless-weather-and weatherless-position-FIXED RATE process
was instituted by some follow-on systems.

As you can see, the battle rages on...

Bob, Wb4APR