[ih] Distributed file systems [was: As Flag Day approaches at CMU]

[Dave Crocker]

On 9/6/2025 5:05 PM, Steve Crocker via Internet-history wrote:
> MIT wasn't the only place where the IMP became the de facto local area
> net.


Whimsically stretching of this model, with the net, itself, being a kind 
of LAN...

At the UCLA Arpanet project, I was responsible for doing documentation.  
But I'd been infected with the Engelbart bug.

So from UCLA, I or a secretary would connect to the ARC system to do the 
document editing.

When done, we'd FTP the file down to USC-ISI, and then run a NetRJE 
program to submit the document to the UCLA Campus Computing machine, 
which had a high-speed, upper/lower case printer.

Side note:  The secretaries complained that the scenario would only work 
about half the time. I turned to my officemate, Jon Postel, to ask 
whether he had any theories.  He made me estimate the component 
reliabilities, where our own system was maybe .8, the net .0, each ARC 
and ISI also maybe 8, and the UCLA campus computer maybe .8.  (I am 
inventing the numbers and not bothering to reconcile them here, so both 
worry about checking them.)

Anyhow, the cumulative probability came out to about .5...

d/

-- 
Dave Crocker

Brandenburg InternetWorking
bbiw.net
bluesky: @dcrocker.bsky.social
mast: @dcrocker at mastodon.social