[ARRL-OK] Packet radio

[Nate Bargmann]

I'll throw my $.02 in since I lived in OK, helped with the nodes and ran
the BBS along with Jeff N5UBY in Enid from about 1993 to 1997.

Looking back over the years it seems to me that the AX.25 protocol was
mis-applied from the start.  What was intended as a means for ensuring
error free transmission for higher level protocols such as TCP/IP soon
became perceived as nothing more than an error free version of RTTY.  It
should have been treated the same as similar link-layer protocols such
as PPP, Ethernet, Token Ring, 802.11a and 802.11b which no one uses as a
TTY link but use as a connection path for higher protocols such as
TCP/IP.

This development in and of itself wasn't bad and it served the ham
community well in the early years.  Later it would become a millstone
around the neck of the experimenters that wanted to do networking and
other such advanced techniques.  Many of these folks just gave up and
migrated to the Internet even though the ability to experiment was near
zero.

Others spent more of their time complaining about the BBS forwarding on
"their keyboard channel" especially once the BBSes started using data
compression and the upper ASCII characters reeked havoc on older
systems.  In short the network was congested and this was due in large
part to inefficient network design since it was difficult enough to
maintain a one lung network into a remote area let alone build the
needed backbone network.

The gear needed to build a high speed backbone was and still is
prohibitively expensive for anything approaching a statewide network.
Site access was becoming harder to obtain by the mid-90s with the rise
of cellular and other commercial communications.  To build a proper high
speed backbone requires nodes every 30 to 40 miles on alternating
frequencies--not a cheap proposition.

Yes, AX.25 has ad-hoc networking built in through digipeating, but once
a certain level of activity is reached, the system grinds to a halt.
The phenomenom is called the "hidded transmitter syndrome" and affects
any well placed station that can hear two or more stations that can't
hear each other.  The result at the central node is a nearly busy
channel due to several stations re-transmitting at once with the central
node left with nearly no chance to transmit at all.  Even with users 
scattered across several local channels the channel loading was severe 
enough that melt-down was achieved and no one got much through.  

Some techniques were tried with varying degrees of success.  One idea
was a centrally located repeater with seperate input and output
frequencies just like a voice repeater.  It came with its own issues as
I'm sure Terry K5HP can describe in detail.

Add to the mix the fact of local Internet access for a low monthly fee,
reliable messages, connectivity to nearly any part of the globe and
packet was abandoned in short order.  A few of us tried to hang on, but
the writing was on the wall and I finally pretty much gave up on packet
around 1999 to 2000 after about 12 fun-filled years on the mode.

All of this said, with the current surplus 1200 bps AX.25 equipment that
is still out there, perhaps the best use for is, ironically enough, as
an error free RTTY system.  APRS stands as really the only made-for
packet application that treats AX.25 as a link layer protocol.  I
understand that APRS has experienced many of the same loading problems
as the early packet network (netwrek if anyone remembers one of my pet
terms).  One difference is that APRS doesn't guarantee error free packet
delivery so some packets can be dropped at the choke point and may get
through later when channel conditions permit.

Yes, my post is quite techinical, but I think packet's demise from the
Ham Internet is as much a technical issue as it is a social one.
Unfortunately, little success was had trying to get disparate groups to
work together to build a reliable network and money was always a major
issue.  It also seems that the rise and death of packet, at least in
some of the more remote areas, caused a change in operating patterns and
voice repeater use fell off markedly as packet took off, but hasn't
recovered in many areas since packet died, which is sad.

73, de Nate >>

-- 
 Wireless | Amateur Radio Station N0NB          | "We have awakened a
 Internet | n0nb@networksplus.net               | sleeping giant and
 Location | Bremen, Kansas USA EM19ov           | have instilled in him
  Amateur radio exams; ham radio; Linux info @  | a terrible resolve".
             http://www.qsl.net/n0nb/           | - Admiral Yamamoto