[ax25-layer2] Some More Food for Thought

[Robert Bruninga]

I'm glad Pete clarified the scope of this discussion, 
and I think it makes sense, because there is a
world of difference between layer-2 connectivity
say on a 300 baud fading Rayleigh HF shared 
channel  and a 9600 baud reliable backbone with only
two TNC's on it.  The AX.25 needs to adjust and
not be too dependent on user settings.

I hope the new additions to AX.25 will allow for backward
compatibility, meaning that an old TNC that tries to
connect to a new one will get a compatible response.
Yet new TNC's that attempt to establish a connection
with a new TNC, will exchange capabilities and
use what is best for that link.   Or something like that.

Bob, WB4APR


>>> pete at ae5pl.net 09/17/05 7:45 AM >>>
My initial reaction to your post was "Yes!".  But then I started to
think about it and the real response is "sometimes".  A standalone
digipeater, for instance, would require capabilities not currently in
most TNCs.  I say most because I know of at least two TNCs which are
incorporating the NSR UI algorithm.  The algorithm is simple enough to
fit in the code space of most if not all TNCs on the market today so
older equipment does not need to be discarded.

The "NIC", on the other hand, can be a KISS TNC.  A good comparison is
the NIC found in most desktop computers today.  Most of these NICs are
basically "dumb".  They process the Ethernet layer and the physical
layer, but they are oblivious to the higher layers such as IP.  They
depend on software drivers to actually handle the layer 3 interface.
This is similar to AGWPE and the Linux kernel AX.25 support working with
KISS TNCs.

You can, however, get a "server" NIC which moves much if not all of that
driver overhead onto the card itself.  This would be equivalent to
having an AX.25 device which provides a layer 3 interface directly.  In
a way, a tracker is such a device since the layer 3 (in this case
OpenTrac or APRS) is contained within the device.

Also, there is nothing in the AX.25 layer 2 specification that restricts
what layer 1 mechanisms can be used.  This knowledge allows for pursuit
of new and different ways to transport AX.25 across the airwaves like
Jeff, et. al. were talking about.  By keeping the focus on layer 2, many
possibilities open up with what can be done at layer 1, layer 3, and
above.

Hope this clarifies things.  This is a work in progress and that is why
this SIG exists: to discuss how to go about implementing these things
and what "things" should or should not be implemented.

73,

Pete Loveall AE5PL 

> -----Original Message-----
> From: James Wagner
> Sent: Friday, September 16, 2005 4:20 PM
> To: Discussion of Link Layer use of AX.25
> Subject: Re: [ax25-layer2] Some More Food for Thought
> 
> I may be missing some things here, not being a networking 
> wonk, but it sound like you are suggesting a replacement for 
> "KISS TNC". That would be a Good Thing, IMHO.
> 
> Jim, KA7EHK

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