On 01/25/2014 09:48 AM, Brad Parker wrote:
On 1/20/14, 12:20 PM, Al Kossow wrote:
Just throwing this out to see what other people
think.
I suspect we're at the tail end of the usage life of devices that
don't speak IP.
I'm mostly thinking about networking devices 80's > 00's
It's a
very interesting question. I'm not sure I can answer without
injecting my own experience and bias.
As to IP going away, IPV6 has the ability to go on (enough bits) and
the basic protocol fits in small things
now and its well known. The embedded developer will only jump when its
cheaper or does something
badly needed.
I'd like to see a smattering of ethernet-ish
non-ip work. Xerox XNS,
Novell Netware, Apple Localtalk, Decnet. Certainly 3Mb and 10Mb
ethernet. And naturally IBM SNA with a 3270 and CICS. A basic
tribute to the phsyical layer, twinax, coax, big ethernet with vampire
taps, twisted pair, fiber.
Something should to be said about IBM token ring. And Apollo's ring.
I think at one time DEC's internal decnet network was pretty damn
big. Worth mentioning DDCMP and phase iv.
By 1983 it was over 5000 nodes on more than 4 continents.
All of the netware and appletalk are in the 1985 time frame. Don't
forget the othe rlike Bayan Vines and a
few more obscure protocols.
Networking is older than that. Look up AlohaNet and UUCP to UUCP
networking.
Some tribute to low speed serial, with current loop
and rs-232
terminals, from the ASR's up to X terminals.
Early modems, including that (swear words) Anderson Jacobson acoustic
coupler and the early Hayes products.
I don't know what to say about ISO protocols.
It might be interesting to say something about military/ga avionics
comm standards. There are a lot of them and some of them spilled
over. I laugh because now days you update your NAV gear with a USB
stick (on the ground :-)
Seems there is confusion over local and wide area networks.
The BBN/MIT/ARPAnet IMP work deserves it's own
space. It relied on
what was around at the time (and yes, how that wikipedia article could
not mention Interdata is beyond me). But it should be noted that TCP
was not the first stream protocol they used.
Maybe a nod to MIT Chaosnet? (ok, Aloha too. perhaps the whole
ethernet thing is it's own space).
Seems some overlap between connectivity and the overlaying protocol
used. Aloha introduced
protocals that could handle the fact that connectivity for its parts
were radio links. Sometime the
media or phyical level drives protocol.
Maybe something about the phone system, T1 lines and
sync framing. How
the T1 evolved from pcm audio to framed data. (where is my
picturephone!!?? :-) ISDN happened along the way, but faded quickly.
Worth mentioning as a false start.
(I know this is all US centric; Thinks like MINITEL also played a part
I'm sure, and 50 other projects I never heard of)
I wonder if a small tribute to UUCP is in order. Which begs the
question of FIDOnet (which I don't know much about). And well,
there's CIS. I once broke into a fleet of dec-10/20's (with a modem)
as a youth using a sysop password someone gave me and marveled at
being able to hop from one host to another - I must have jumped
through 20-30 machine before I got bored. I'm guessing that was all
decnet-over-sync-lines.
What about amateur radio? Hams had been doing FAX/slowscan images and
RTTY in the past and also have in the last
several decades packet radio to the current "sound card modes" Where
the PC and a sound card encode and decode
multiple different encodings and protocols. I will include D-Star which
is both text and speech over mixed media
(hamd radio and internet linking).
(that was a fun cat and mouse - it was clear the owner
was running
something in the background which was logging activity. i discovered
the logs and then hacked the monitor program not to log and erased the
logs. he kept 'fixing it' and restarting it and I kept 'unfixing it'
and erasing. that went on for days. i hope he could see the humor in
it all. so much for a misspent youth.)
Wow. Yes, daunting. Best of luck.
Very daunting to even cover a slice of that.
Allison