[aprssig] are write-only APRS-IS clients valid?

[John Gorkos]

>
> You are correct, you don't understand how an APRS-IS server dupe elimination works or how a limited feed from a server works.  Bottom line: if the server that the bidirectional IGate is connected to sees the packet from the RX-only IGate before it sees the packet from the bidirectional IGate and the bidirectional IGate's server is connected to a filtered port (common combined server/IGate configuration), the upstream server will not pass messages destined for the station the bidirectional IGate just tried to gate.  Messaging is busted in either case with a RX-only IGate and the proverbial "I never see an ack but I can see my packets on aprs.fi" complaint is made.
>
> Again, I have explained it many times and if you still don't understand how dupe elimination in servers can affect IGates receiving limited feeds from the servers, I recommend you visit www.aprs-is.net and learn more about the modern-day network.
>
> Again, lack of understanding of modern-day clients.  It is just as easy to click to boxes to make that client gate to/from RF at all.  Most GUI clients allow connection to RF and to APRS-IS without gating between them.  Not a good reason to run an IGate that if left on due to an evacuation could further break the emergency communications you want it to support.  Best to be not IGating at all than be a receive-only IGate (SatGates are not included in this statement).
>
> 73,
>
> Pete Loveall AE5PL
> pete at ae5pl dot net
>
>
Gee, Pete, if you would perhaps release the code that actually does 
this, in a public manner, we could all learn from you.  It seems like 
the aprsc guys have it figured out, sorta, but since your code is the 
"canonical" way to do things, perhaps it would be immensely helpful to 
the community if you'd pick one of the many licenses that allow you to 
share your code without fear of it being "ripped off" and used by 
someone else to make money.
I'm not a great Java programmer, so I'd love to see how your code works. 
I might learn something.  It would also give you a great deal of 
leverage to be able finish arguments like the one in this thread by 
saying "go look at the code, that's how it works."

Seriously, look around.  The Europeans have all embraced an open source 
model, Andrew's done it with YAAC, my libraries and web sites are all 
published on github, Bob made his DOS code available.  What are you 
gaining by taking your work to the grave with you?  What are you 
preventing, other than a the free exchange of ideas in a public service 
community, and ability of future coders to build on past success instead 
of continually reinventing the wheel?

Let us see how you do it, so we can make sure we're all doing it 
correctly, or at least have a common-ground conversation about what 
"correct" is.

John Gorkos
AB0OO