16 Sep
2007
16 Sep
'07
6:10 p.m.
I don't see why not; serial comms (over whatever medium) connecting two
machines is still 'networking' in my book. I the case you mention it may well
be only point-to-point rather than "lots of machines on the same line", but I
still think it qualifies.
Then let's go a step further. When/where did the first internetworking
gear come onto the market? Not multi-protocol routing, just routing in
general: segmented networks with nodes that moved traffic towards remote
nodes on the behalf of local ones.
16 Sep
16 Sep
6:50 p.m.
On 16 Sep 2007 at 13:10, ajones wrote:
Then let's go a step further. When/where did the
first internetworking
gear come onto the market? Not multi-protocol routing, just routing in
general: segmented networks with nodes that moved traffic towards remote
nodes on the behalf of local ones.
Does Usenet UUCP-type networking count?
Cheers,
Chuck
17 Sep
17 Sep
1:56 a.m.
Then let's go a step further. When/where did the
first internetworking
gear come onto the market? Not multi-protocol routing, just routing in
general: segmented networks with nodes that moved traffic towards remote
nodes on the behalf of local ones.
Ask Western Union and Kleinschmidt.
--
Will
4:03 a.m.
On 16 Sep 2007 at 20:56, William Donzelli wrote:
Or perhaps earlier, when telegraph operators simply retransmitted the
copy. If "unattented" is the operative answer, then Edison in 1864
with his "repeating telegraph".
If it's not electrical, how about semaphores, smoke signals, flags,
laterns, drums, whistles, riders and runners?
Cheers,
Chuck
Then let's
go a step further. When/where did the first internetworking
gear come onto the market? Not multi-protocol routing, just routing in
general: segmented networks with nodes that moved traffic towards remote
nodes on the behalf of local ones.
Ask Western Union and Kleinschmidt.
1:40 p.m.
Or perhaps earlier, when telegraph operators simply
retransmitted the
copy. If "unattented" is the operative answer, then Edison in 1864
with his "repeating telegraph".
If it's not electrical, how about semaphores, smoke signals, flags,
laterns, drums, whistles, riders and runners?
I should have said "Ma Bell", now that I think of it. Pretty much all
of basic concepts of computer networking* can be found in Notes on the
Network. All that fancy stuff about routing, least cost methods,
network failures, hierarchies, network addressing? All in there from
way back.
And of course, the telephone network is still the largest.
* ARPAnet, for getting the lions share of credit for inventing the
computer network, mostly just invented (and implemented) packet
switching.
--
Will
1:59 p.m.
* ARPAnet, for getting the lions share of credit for
inventing the
computer network, mostly just invented (and implemented) packet
switching.
Oops, back up...leave out "invented". I forgot there were a few
interesting packet switched network before then. Sorry, I am rushed
today...
--
Will
5:24 p.m.
"William Donzelli" wrote:
* ARPAnet, for getting the lions share of credit for inventing the
computer network, mostly just invented (and implemented) packet
switching.
This is off topic, and we should stop, but I don't think this accurate.
Technically the phone system is circuit switched. The ARPAnet did
pioneer the notion of concatenated networks, or a "catanet" (if I recall
correctly) which was dynamically reconfigurable. The original idea was
to create a network which would survive a nuclear blast, which the phone
system would not, at least not it the sense of being dynamically
reconfigurable.
These days the internet is pretty statically routed, since BGP setup is,
well, more of an art than a science and peering is more political than
technical.
I think hop by hop routing with dynamic reconfiguration was bit of an
innovation.
-brad
19 Sep
19 Sep
6:11 p.m.
This is off topic, and we should stop, but I don't
think this accurate.
Network history off-topic? WTF?
Technically the phone system is circuit switched. The
ARPAnet did
pioneer the notion of concatenated networks, or a "catanet" (if I recall
correctly) which was dynamically reconfigurable. The original idea was
to create a network which would survive a nuclear blast, which the phone
system would not, at least not it the sense of being dynamically
reconfigurable.
Yes, true - if a line went down - hang up and try again!
I think hop by hop routing with dynamic
reconfiguration was bit of an
innovation.
Very much so - I do not doubt that. However, all that hop-by-hop and
reconfiguration stuff was built on Ma Bell's work. Hop by hop? Well,
that can be found in the section about how a call is routed from the
local exchange, up the hierarchy to the long-lines people, and back
down to another local exchange. In a "complex" call, like Ichibutt, NC
to Bumbledweeb, ME, there may be four or five hops needed to complete
the call. And the reconfiguration? Essentially grabbing on to the ever
changing pool of free trunks and long lines.
--
Will
9:25 p.m.
On Wednesday 19 September 2007 13:11, William Donzelli wrote:
I never could understand how they managed to do all of that with what was
basically electromechanical stuff...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
I think hop by
hop routing with dynamic reconfiguration was bit of an
innovation.
Very much so - I do not doubt that. However, all that hop-by-hop and
reconfiguration stuff was built on Ma Bell's work. Hop by hop? Well,
that can be found in the section about how a call is routed from the
local exchange, up the hierarchy to the long-lines people, and back
down to another local exchange. In a "complex" call, like Ichibutt, NC
to Bumbledweeb, ME, there may be four or five hops needed to complete
the call. And the reconfiguration? Essentially grabbing on to the ever
changing pool of free trunks and long lines. 
10:01 p.m.
Roy J. Tellason wrote:
I never could understand how they managed to do all of
that with what was
basically electromechanical stuff...
Some very tricky relays. That is why you have 1 for long distance.
I had a book about that -- 1940's ( but it got lost ) and the big
thing was automatic dialing. I am not sure but I think on the old non dial phones
you could even back then dial your number by tapping the hook on the phone.
Ben.
10:12 p.m.
On Sep 19, 2007, at 5:01 PM, woodelf wrote:
You could; I did it as a child.
-Dave
--
Dave McGuire
Port Charlotte, FL
Farewell Ophelia, 9/22/1991 - 7/25/2007
I never could
understand how they managed to do all of that with
what was basically electromechanical stuff...
Some very tricky relays. That is why you have 1 for long distance.
I had a book about that -- 1940's ( but it got lost ) and the big
thing was automatic dialing. I am not sure but I think on the old
non dial phones
you could even back then dial your number by tapping the hook on
the phone. 
3:03 a.m.
"William Donzelli" wrote:
...
the call. And the reconfiguration? Essentially grabbing
on to the ever
changing pool of free trunks and long lines.
I would argue that the routing was (at that time) essentially static. As I
understand it CO's were not able to dynamically reroute calls if a trunk
failed - that had to be configured in as an alternate path.
The catanet model would allow 75% of the network to go away and it would
dynamically learn a new way to route to the destination; and the fact
that it was packet based and not circuit swithed would mean that it was
self healing. Not possible with circuit switches; you have to hang up
and redial and hope that an alternate path had been configured in.
-brad
17 Sep
17 Sep
4:09 a.m.
I forgot pigeons.
Here's a good source:
http://labit501.upct.es/ips/libros/TEHODN/toc.html
Cheers,
Chuck
4:16 a.m.
Then let's go a step further. When/where did the
first
internetworking gear come onto the market? Not multi-protocol
routing, just routing in general: segmented networks with nodes that
moved traffic towards remote nodes on the behalf of local ones.
My first reaction was "doesn't UUCP fit?", as someone else alluded to.
But really, doesn't the postal system fit, with traffic being mail and
nodes being sorting offices? Or, for that matter, any large shipping
concern, such as the Dutch, the Spanish, the English, at suitable
points in history?
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents.montreal.qc.ca
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
6327
days inactive
6331
days old
14 comments
9 participants
participants (9)
-
aijones2＠bsu.edu -
alexandre-listas＠e-secure.com.br -
bfranchuk＠jetnet.ab.ca -
brad＠heeltoe.com -
cclist＠sydex.com -
mcguire＠neurotica.com -
mouse＠Rodents.Montreal.QC.CA -
rtellason＠verizon.net -
wdonzelli＠gmail.com