Linux-Hams archive - April 1997: Re: AX.25 references/documentation

Re: AX.25 references/documentation

Terry Dawson ()
Thu, 10 Apr 1997 17:48:15 +1000 (EST)


Mike Bilow warned;
> > Above all, don't make the
> > mistake of assuming that AX.25 is just a ham radio flavor of X.25.

Ramon replied:
> I don't completely agree. AX.25 is heavily based on X.25, where the address
> field and the use of repeaters is extended. But the signalling (SABM, RR,
> DISC, etc) is quite similar to X.25. AX.25 is closely related to X.25, but
> there are HAM-specific differences.

AX.25 is based heavily on the datalink protocol specified within X.25,
which is based on HDLC. Most of the X.25 specification is completely
irrelevant to AX.25.

AX.25 + Rose is much more closely related to X.25, but there are still
some significant differences.

> Anyway, how official is AX.25? The X-protocols are developed/approved by
> ANSI committees if I'm not mistaken, who is the regulatory organ that
> gives/gave the advisory on AX.25 ?

How official does it need to be ? There is a published specification backed
and supported by a demographic comprising the community that use it. Sounds
official to me.

X.25 is a C.C.I.T.T. recommendation. ANSI have nothing to do with it.
I think ISO-8203(?) is the ISO standard.

> (Please don't understand this wrong... the HAM radio world basically agrees
> on WHAT AX.25 is, and what its range of use is.)

Actually, recent discussion leads me to eblieve this isn't actually
very true. There are a number of 'extensions' to AX.25 that are not
'official' in the same way the AX.25 specification is. It would be
a profitable and useful exercise I believe to at least collect and
document these in an attempt to standardise them.

> Makes us think about better protocols than AX.25 with HDLC. Something that
> takes care about error-avoidance and recovery in a radio environment. This
> is something where we (the radio amateurs) can contribute to, something that
> hasn't been chewed on by the academic and industrial world for years and
> years.

Yep it's time we started.

> And long live Linux! At least an operating system that allows us an easy
> implementation of whatever protocol we want to experiment with... The times
> of NOS and MS/DOS are gone...

not gone, but fading perhaps. They still have a place.

regards
Terry