| ||||||||
Election 2006 (and beyond): Digital Copyright Canada
From: russell_-at-_flora.ottawa.on.ca
Date: 28 Oct 1998 09:05:56 -0500
Every "real life story" I have ever heard amounts to bad management
decisions, and a breakdown of the points I raised in my paper:
"Computer Software is like any other information used by humans or
machines and needs to be:
- periodically updated as conditions change
- thoroughly tested (peer-reviewed, verified, tested in many
environments, etc)
- not be relied upon unless the last two conditions are met.
http://www.flora.org/russell/work/y2k.html
With the average business-person only thinking one or two quarters away,
and governments barely thinking until the next election, any suggestion
that these groups will be able to help us out of any potential problems is
extremely unlikely (And I would suggest dangerous).
Corporations and governments are now, under the pretense of "Fiscal
responsibility", externalizing pretty much any expense that they feel they
can shove under the rug. Lack of updating of critical information (Not
just software), and forcing reliance on unreliable systems (through
privatization/etc to even less long-term accountable institutions) are
just parts of this puzzle. Y2K is just one of millions of such problems
that happen when this form of irresponsibility requires immediate payment.
It is important to watch how this unfolds. Terry's recent pointing to
a Globe&Mail article titled "Army fears civil chaos from millennium bug."
<http://www.inode.org/y2k-archive/msg00032.html> and his forwarding of
Robert Theobald's commentary is timely. An important quote from the G&M:
The success of the operation depends on "public confidence in the
government's ability to manage and provide leadership in dealing with
the year 2000 problem," the order says.
That makes me worried as they can't quite be expected to provide
leadership on an issue that exists precisely because of their lack of
leadership and long-term-thinking in the first place! If there are
failures due to Y2K is is not because Y2K is a "big problem", but because
the failed systems were not properly maintained and inappropriately
brittle to begin with.
---
Russell McOrmond, Internet Consultant: <http://www.flora.org/russell/work/>
http://www.flora.org/russell/work/closed-ms1.html Is MS a Monopoly?
CKCU Funding drive, FLORA, Y2K http://www.flora.org/?ckcu
---------- Forwarded message ----------
Date: Wed, 28 Oct 1998 05:01:53 -0800 (PST)
From: Declan McCullagh <declan@well.com>
To: politech@vorlon.mit.edu
Subject: FC: One company's real life Y2K problems
---------- Forwarded message ----------
Date: 28 Oct 1998 04:17:45 GMT
From: Blake Leverett <bleverett@nospam.worldnet.att.net>
Subject: Real Life Y2K failure
Here is a real-world example of a Y2K failure. A buddy of mine (call him
"Jim") is the network administrator at a small (8 Mill/yr sales) company. I
can't say the company name, so if you are going to whine about how I am
making this up, don't bother reading on. But I am familiar enough with the
situation (I know some other people there) to know that he is not making
this up. It is a manufacturing company.
"Jim" warned the company management over a year ago that the company's
software would self-destruct in October of 1998 because of the year 2000
bug. It looks out 15 months into the future and although the software is
configurable, it can't be configured to look ahead for a shorter time
period. The software is an old DOS package, and it's not Y2K compliant. It
runs everything from manufacturing documentation to accounting, payroll and
accounts payable included.
The company management dragged its feet until March. Finally, after "Jim"
threatened to quit, they decided to shell out the $100K or so it cost for
new software. Months of work followed by both the software vendor and "Jim"
to make the switchover happen. Well, here it is the end of October, and the
new system isn't up yet. They started too late, so the old software is
still running, and is peeking over the Y2K boundary.
There have been problems. The database files get scrambled on a daily
basis. "Jim" has to edit these files manually with a hex editor to fix
them. The headers get screwed up, and he uses a hex editor to side-by-side
compare the old (good) file with the new (screwed) file. Then he guesses
what to edit to make it limp along for another 4 hours. Sometimes the files
can't be repaired, and the lost data must be re-entered from paper copies
(if any exist).
The real kicker came the other day when they ran Accounts Payable. They are
a wee bit behind on their AP, so they paid out partial payments to many of
their vendors. A total of $150K worth of checks was printed, and promptly
mailed. Shortly thereafter, the system crashed yet again. The record of
the latest checks was lost. Since a large number of vendors were paid, they
now have no idea of what they owe to any vendor. It effectively destroyed
all records of their AP.
Y2K is serious. This little manufacturer is wasting lots of time fighting
Y2K problems (100 man-hours this month so far), and is losing money from
lost information. And they are in good shape - my friend "Jim" saw this
coming and insisted that the problem be fixed at great expense to the
company. He is a big asset to the company, so they listened when he
threatened to quit. If he were mediocre, they probably would have called
his bluff. They will be ready (supposedly) on November 1st. But there are
THOUSANDS of similar small companies that haven't shelled out big bucks for
a fix. They will start seeing problems soon. And the problems will get
worse and worse as more routines look past 1/1/00.
Although this company is experiencing lots of computer problems, the
customers have yet to find out. The customers can only interact with the
company via telephone with the sales or service departments (or www), so
internal problems are easily concealed. But their efficiency and
effectiveness has been reduced significantly. And it has cost them dearly
both to fix the problem and to not have it fixed in time.
I hope this provides a small example of what Y2K can do. Computers are
boring and unimportant devices - until they quit working. There are very
few companies that can survive for long without their data.
Blake Leverett
--------------------------------------------------------------------------
POLITECH -- the moderated mailing list of politics and technology
To subscribe: send a message to majordomo@vorlon.mit.edu with this text:
subscribe politech
More information is at http://www.well.com/~declan/politech/
--------------------------------------------------------------------------
| Please read the FLORA.org Terms and Conditions before you submit information to FLORA.org | |
|
(USA) (Canada) |
|