X-Message-Number: 10115
Date: Sat, 25 Jul 1998 09:49:14 -0700
From: Peter Merel <>
Subject: Y2K URLs and Replies

The range of responses yesterday was pretty representative of what folks 
are saying in other parts of the net. Many folk feel it's Chicken Little to
be worrying about such things now. Many think it's just a software problem
that will be fixed before it can occur. Many feel it'll be the cause of 
some disruption, but nothing that'll knock society off its pins. And many 
feel it's going to be devastating.

I feel you have to make up your mind based on facts. Having described 
a worst-case, what I'd like to do is develop sufficient understanding of 
the issues to consider the odds of such events in an informed way. To 
that end, here's some URLs:

http://www.accsyst.com/writers/embedded.htm
http://www.businesstoday.com/techpages/y2kindex03.htm
http://www.sustainableworld.com/y2kgps/gpsbug.html
http://catless.ncl.ac.uk/Risks/18.96.html#subj9
http://www.garynorth.com/y2k/detail_.cfm/1163


http://www.cs.nps.navy.mil/people/faculty/mjholden/courses/csr100/Y2Kslides/tsld009.htm

--

To specific points:

Perry Metzger writes,

>Where are you getting all this from? Are you just making it up?

Hell no. There's no end of paranoia online - I could cut and paste worse
stuff than what I wrote until the cows come home. For example, here's a 
long page of banking-specific links: 

http://www.garynorth.com/y2k/results_.cfm/Banking

[on electrical disruptions on 1/1/00]

>How? Where do you think the average electrical generator stores the date?

Check the embedded systems link above. Worth noting that yesterday Sweden
announced it is preparing to shut down all its nuclear reactors around 1/1/00
for this reason. In a nutshell, it looks like somewhere between 5% and 20% 
of the PLCs and control points in the world's electrical generation and
distribution facilities may fail. What the cumulative effect will be is
unknown.

[telecommunications]

>SS7 doesn't even have a year field in it.

Embedded systems here. PLCs, RTUs, SNMP and CMIP boxes manage all telecoms,
and many have vulnerabilities. Many PABXes do too. The installed base is 
vast, and since we're talking about firware there's nothing to do but
replace it.

--

Linda Chamberlain writes

>Create liquidity (three month s cash on hand, advance payments on 
>monthly bills, traveler s checks).

In the event of a financial crash we might see either dramatic deflation
or inflation; so it might be good to also have some gold coins and barter
goods at the ready too. And try to get notarized paper records of all 
your accounts.

--

Bob Ettinger writes,

> Ways will be found. Human stupidity is formidable, but there are limits.

I'd feel a lot more comfortable about this if we had several thousand
times more people interested in cryonics than we do right now. If we 
learn anything from the last few decades' reaction to your vision
of emortality for all, it's that where human stupidity is concerned 
all bets are off.

Major stupidities I see going on at the moment are:

* Inadequate awareness of the scope of the problem. About scale of resolving
the embedded systems issues the US Navy report above says, "Imagine replacing 
the entire interstate road system ...".

* Most management thinks this is a software problem. While software is sure
involved, for the most part it's actually a firmware and economic problem.

* Inadequate resolve to fix the problem in non-US countries. For example,
where most US banks have budgeted $200-$600 million for fixes, a recent
survey found that the Japanese banking system _in its entirety_ has allocated
less than $300 million to it.

* A prevailing attitude in many industries of "fix on failure". 

* Almost zero recognition of the problems of ripple and synchronicity 
of failures. For some examples of such snafus, see 
http://millennia-bcs.com/jonexist.htm

* No recognition that most non-US countries have imported technology but
not expertise; many have nowhere near the technical wherewithall to do 
anything about it even if they were properly aware of it. Which they're not. 

* Most engineers are and would rather be developing new systems than testing
and replacing legacy firmware.

>a power company could [instead of] computer-controlled decisions [...]
>make hourly decisions based on averages.

Sure it could, if it could find and replace each dud PLC in its 
distribution network.

> [government and business should] just send out average payments to
>the usual people for a while. And so on and on. Calm and common sense.

I sure wish I lived on a planet where calm and common sense were the usual
response to mass economic disruption. Unfortunately, the only examples
of mass economic disruption history teaches me speak of great panic and 
irrationality. If you can think of good reasons why this time will be 
an exception to this I'll be very much relieved!

[a bank run early to mid 1999] 

>Early 1999 is only around 6 months away. One obvious recourse for 
>the depositor is to ascertain, very soon, whether his own bank is 
>internally vulnerable to the y2k problem.

Bank runs don't come from the y2k problem itself; they come from the
preparations that folk will make to ensure their liquidity over y2k. Linda
Chamberlain described these - they begin with "take out 2-3 months cash". 
When everyone does that at once the supply of banknotes dries up and panic 
and domino effects do the rest. 

Just when this happens depends on when people start to worry about
it happening - I've already seen newspaper reports on it, so I wouldn't 
be surprised to see it early 1999. Given the stupid human tendency to go
from apathy to panic overnight mid 1999 seems more likely to me.

Peter Merel.

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=10115