[ax25-layer2] IP Routing

[Robert Bruninga]

>>> pete at ae5pl.net 09/20/05 11:13 AM >>>
>> ... any change... should result in a system 
>> that is at least as robust as the current system...
>
>This "robust" system has all but disappeared from 
>the US (except forAPRS and some Winlink Paclink 
>usage).  Why?  Because it was just that "robustness" 
>attempted to used at the layer 2 which doomed it.

No way.

1) What doomed "packet radio" was the explosion of
the internet and 56 kB modems in the mid 90's.  1200
baud packet radio could not compete.

2) For those of us that were there, the HAM AX.25
network at THAT TIME had evolved into a 99%
layer 3/4 system of NODES and digipeating was
absolutely eliminated.    Of the hundreds of level
4 nodes that supported all of the AX.25 system all
over the country,  it was imposible to find ANY that 
would support even a single digipeat hop.

I konw this well, because that was when we began
APRS and it was impossible to do anything but
connected packet via a level 4 node.  That is why
APRS had to start all over from zero and implement
digipeaters one at a time on a new frequency.

So, dont tell me AX.25 is broke because of digipeating.
It had evolved very well into a level 4 network with global
connectivity and with zero digipeating.

What killed it was the 56k global bandwidth of the
internet available to anyone and everyone. (now
most of us have megabit access)...  that said,
it would be dreaming to think that forceably killing
the user-selected-routing option would bring it all
back...

Yes, the node system was still source routing
at its worst, and that is worthy of fixing, but be
very careful where you place the blame.   If
NSR is better, then people will use it.  If it is not
better, then they should have the option of not
using it within the published protocol.

Bob, WB4APR


  We have
had 20+ failed years of a "strict peer-to-peer topology".  Why do you
think the future holds anything more?  The reason for this SIG is to
find ways to break out of those non-functional modes, take a fresh look
at how to use AX.25 as a layer 2 protocol in a network, and create a RF
LAN basis that amateur networking can utilize as _one_ of the layer 2
protocols.

> Given all of this, I'm not sure that there is a lot we can do 
> to improve layer-2 beyond what is already in KISS. Flow 
> control, while a welcome addition, is really a layer-1 issue. 

No, it would be layer 1 if it were on the other side of the protocol.
This is simply adding inter-layer flow control to the layer 2/layer 3
interface.  This is not only supported but implemented with many other
protocols.

> Adding it is simply a matter of updating the serial port 
> driver in the KISS implementation. 
> KISS that doesn't use a physical serial port doesn't have 
> flow control as an issue - which is why flow control isn't a 
> layer-2 issue.

Actually, KISS still does have a flow control issue, even when
implemented in sound card drivers, etc.  Again, this is a layer 2/layer
3 interface issue, not a layer 1 issue.

Hope this helps clarify things.

73,

Pete Loveall AE5PL 

_______________________________________________
ax25-layer2 mailing list
ax25-layer2 at lists.tapr.org 
https://www.tapr.org/cgi-bin/mailman/listinfo/ax25-layer2