[UFO Chicago] UFO This Thursday

[Jordan Bettis]

On Wed, Jun 25, 2008 at 10:13:41PM -0400, Brian Sobolak wrote:
> 
> 8pm, 4229 W. Irving Park @ Golden Nugget House of W0nderful
> 
> We'll talk about Linux, ASSP, email, coffee, and probably many
> other things.
> 

For the post meeting: 

The book I read about BBN and the IMP was "Computing in the Middle
Ages" by Severo Ornstein, who was an engineer at the company on that
project. The IMP was made out of a Honeywell 516 minicomputer, which
was chosen because it had a reliable service history, there was a
hardened Military version of it (it *was* an ARPA project), and
Honeywell was willing to modify it to fit BBN's purposes. It also came
stock with two communications channels, which they needed to connect
to the host mainframe on one end and the ARPANet on the other.

BBN began writing the software after having only received one 516
machine, and testing the network code by connecting the two
communication channels together. While doing that they discovered the
infamous metastable flip-flop glitch.

<http://en.wikipedia.org/wiki/Flip-flop_%28electronics%29#Timing_and_metastability>

With respect to the polemic about the space truck that couldn't: 

<http://www.washingtonmonthly.com/features/2001/8004.easterbrook-fulltext.html>

I found this article to be excellent. It was written in 1980 before
the shuttle had ever flown and correctly predicted that the shuttle
was a solution in search of a problem, and not a particularly good one
at that. It also identified three big flaws in the machine: The solid
rocket boosters, the fragile heat shield, and the main engines. Two of
those have now killed shuttle crews. I've heard that the main engines
have improved considerably so perhaps the criticisms aren't so valid,
but my money is still on them as the cause of the next roman candle.

-- 
Jordan Bettis