<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<body style="font-family: 'Times New Roman', Times, 'Liberation Serif', serif;">
<div dir="auto">
<div id="aqm-original" style="color: black; font-family: sans-serif;">
<!-- body start -->
<div class="aqm-original-body" style="color: #000000; background: #ffffff;">
<div style="color: black;">
<p style="color: black; font-size: 10pt; font-family: sans-serif; margin: 8pt 0;">On February 14, 2019 10:32:16 Miles Fidelman <mfidelman@meetinghouse.net> wrote:</p><blockquote type="cite" class="gmail_quote" style="margin: 0 0 0 0.75ex; border-left: 1px solid #808080; padding-left: 0.75ex;">
<p>On 2/14/19 9:28 AM, Bernie Cosell wrote:</p><blockquote type="cite" cite="mid:168ec6900b8.2796.742cd0bcba90c1f7f640db99bf6503c5@fantasyfarm.com"><div dir="auto">
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;">my take on that
is that there were two lines of thought leading up to the</div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;">ARPAnet. very
very roughly: one was paul baran's, who was thinking</div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;">about how the
military <span style="font-size: 12pt;">command and control
might be able to continue functioning in the event of </span><span style="font-size: 12pt;">an attack, and JCR Licklider, who
was thinking</span></div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;"><span style="font-size: 12pt;">about how wide-spread researchers
could share resources, ideas and results </span></div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;"><span style="font-size: 12pt;">to better collaborate.</span></div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;"><span style="font-size: 12pt;"><br>
</span></div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;"><span style="font-size: 12pt;">when the ARPAnet got funded by the
DoD, Baran's story was the easier to</span></div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;"><span style="font-size: 12pt;">understand to the average person,
raather than the more diaphanous idea</span></div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;"><span style="font-size: 12pt;">of researcher collaboration. so
Baran's take kinda caught the public</span></div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;"><span style="font-size: 12pt;">imagination, but the reality for
those of us working on it was the it was</span></div>
<div dir="auto" style="font-family: "Times New Roman",
Times, "Liberation Serif", serif;"><span style="font-size: 12pt;">{somehow :o)} to be a research
tool. </span></div>
<br>
</div>
</blockquote>
<p>You were involved a lot earlier than I was. Perhaps you could
comment on how much folks thought about fault-tolerance in the
early days. It's always struck me that things like
continuity-of-operations, in the face of node & link outages,
and no-single-point-of-failure, were baked in from the beginning.
You know - all the stuff that would allow the net to survive
everything from backhoes to natural disasters, and coincidentally,
nuclear war. </p></blockquote></div></div></div><div dir="auto">we couldn't do anything, of course, about continuity of operations, per se, </div><div dir="auto">but we certainly focused on having the IMP be as resilient {including such</div><div dir="auto">hacks as rebooting froma neighbor, rather than someone feeding paper tape</div><div dir="auto">in} . W<span style="font-size: 12pt;">ill Crowther, in particular, wanted the network to </span><i style="font-size: 12pt;">work. </i><span style="font-size: 12pt;"> if there was</span></div><div dir="auto"><span style="font-size: 12pt;">a way {looking from a god's eye view of the topology of the net} to get from </span></div><div dir="auto"><span style="font-size: 12pt;">A to B, regardless of outages and such, we wanted </span><span style="font-size: 12pt;">an isolated IMP, without benefit</span></div><div dir="auto"><span style="font-size: 12pt;">of knowing the topology of the net or a central </span><span style="font-size: 12pt;">oracle telling it how to get where, </span></div><div dir="auto"><span style="font-size: 12pt;">to figure out </span><i style="font-size: 12pt;">if </i><span style="font-size: 12pt;">a rute was possible </span><span style="font-size: 12pt;">and then make an "educated guess" which</span></div><div dir="auto"><span style="font-size: 12pt;">of its modem lines would get </span><span style="font-size: 12pt;">the packet there most efficiently. </span></div><div dir="auto"><span style="font-size: 12pt;"><br></span></div><div dir="auto"><span style="font-size: 12pt;">of course we saw instances of "rerouting" around problems early on. , but the</span></div><div dir="auto"><span style="font-size: 12pt;">two things </span><span style="font-size: 12pt;">i thought were cool was when/if the net got partitioned, the two {or</span></div><div dir="auto"><span style="font-size: 12pt;">more} </span><span style="font-size: 12pt;">sections just kept going and then quietly stitched itself back together when</span></div><div dir="auto"><span style="font-size: 12pt;">the necessary links were restored, and when a new IMP came online <i>nothing </i></span></div><div dir="auto"><span style="font-size: 12pt;">had to be done. you just turned it on and it, without further ado, it said hello</span></div><div dir="auto"><span style="font-size: 12pt;">to its neighbors and became part of the mix. </span></div><div dir="auto"><span style="font-size: 12pt;"><br></span></div><div dir="auto"><span style="font-size: 12pt;">i think we didn't fully appreciate at the time nust how resilient the damn</span></div><div dir="auto"><span style="font-size: 12pt;">things were and how that shaped the perception of it. very quickly, scarily </span></div><div dir="auto"><span style="font-size: 12pt;">so if you knew about the insides of the IMP 😊, the network became almost</span></div><div dir="auto"><span style="font-size: 12pt;">a utility. Len at UCLA and folks at BBN discovered that the idea of the ARPAnet</span></div><div dir="auto"><span style="font-size: 12pt;">as being a <i>networking e</i>xperiment got hamstrung. it had kinda showed</span></div><div dir="auto"><span style="font-size: 12pt;">that the network level questions had been answered - routing worked, the</span></div><div dir="auto"><span style="font-size: 12pt;">network was resilient, rhe queuing theory worked, heterogeneity worked. </span></div><div dir="auto"><span style="font-size: 12pt;">of course none of it was perfect, but it <i>worked. </i>and that almost immediately </span></div><div dir="auto"><span style="font-size: 12pt;">changed the "conversation".. there was now a working network and while the</span></div><div dir="auto"><span style="font-size: 12pt;">squirrels kept spinning the exercise wheel to improve the infrastructure </span></div><div dir="auto"><span style="font-size: 12pt;">the focus shifted to question of whar to do with the damn thing.</span></div><div id="aqm-original" style="color: black; font-family: sans-serif;" dir="auto"><div class="aqm-original-body" style="color: #000000; background: #ffffff;"><div style="color: black;"><blockquote type="cite" class="gmail_quote" style="margin: 0 0 0 0.75ex; border-left: 1px solid #808080; padding-left: 0.75ex;">
<p>On the physical side, the early IMPs were pretty rugged boxes
(not so much C/30s and such). Were any of the IMPs built to
withstand EMP?</p></blockquote></div></div></div><div dir="auto">i was told the the original 516s were fully ruggedized and tempest secure. </div><div dir="auto">dunno about EMP. i was just amused by this big, solid monster.. it seemed</div><div dir="auto">overkill. but at least one thing that accomplished was to keep curious </div><div dir="auto">pokers at the sites from messing with it. i reall don't think "war survival" </div><div dir="auto">of the system was much of a consideration. perhaps dave remembers </div><div dir="auto">how they decided on the full-military cabinet. i joined the <span style="font-size: 12pt;">project after it was</span></div><div dir="auto"><span style="font-size: 12pt;">a fait accompli. </span></div><div dir="auto"><br></div><div dir="auto">/b\</div><div dir="auto"><br></div>
<div id="aqm-signature" dir="auto" style="color: black; font-family: "Times New Roman", Times, "Liberation Serif", serif;"><div style="margin: 0; color: black; font-family: 'Times New Roman', Times, 'Liberation Serif', serif;"><br></div><div style="margin: 0; color: black; font-family: 'Times New Roman', Times, 'Liberation Serif', serif;"> Bernie Cosell</div><div style="margin: 0; color: black; font-family: 'Times New Roman', Times, 'Liberation Serif', serif;"> bernie@fantasyfarm. com</div><div style="margin: 0; color: black; font-family: 'Times New Roman', Times, 'Liberation Serif', serif;">— Too many people, too few sheep —</div></div>
</div></body>
</html>