I read www.garynorth.com regularly. It is a good source of information,
but you have to take it with a grain of salt, realize that he is filtering
the data and may be presenting it in misleading ways. Someone pointed
for example to http://www.garynorth.com/y2k/detail_.cfm/1260, which
North characterizes as a banker testifying about the threat caused by
the collapse of the telecommunications system. If you follow the link
to the original talk at http://www.house.gov/banking/2598acb.htm you
will see that North provided an out of context quote which was inserted
as a caveat in a talk which otherwise discussed positive efforts to
address Y2K problems. The speaker has no particular knowledge about
telecommunications (he is addressing banking issues) and was simply
pointing out that the banks were relying on the phone system continuing
to work. If all you did was read North's synopsis you would have
come away with a false impression of the thrust of the talk. So it's
important to follow the pointers and not just read North's excerpts,
and also to realize that North is going to focus on bad news.
> Another concern was the nature of the posters on certain y2k forums. Try
> comp.software.year-2000. A large number of very experienced mainframe
> and other programmers, many working on y2k issues, post there.
> Admittedly, you also get Paul Milne, a non programmer who paints some of
> the scariest post y2k scenarios, and is annoyingly christian (though
> again, not of the millenialist variety).
This is a much more balanced forum than North's and Yourdon's BBS's. You
will find many experienced engineers who can put the problems into
perspective.
> And Hal also took a cheap shot at Ed Yourdon. Check out his
> bio(somewhere on www.yourdon.com) and you'll see he has quite the
> reputation to risk. Is he supposed to work for free?
Yourdon is selling his book and his Y2K consulting services. He has a
financial interest in making the problem look bad. No one is going to
buy the book and consulting if he tells them that Y2K probably isn't
going to be much of an issue. You have to keep this kind of thing in
mind when you read his stuff. Also take a look at
http://www.albany.net/~dmills/fallback/fintro.htm (follow the links at
the bottom) where a power engineer criticizes some of the claims made
in the Yourdons' book.
> For a view from
> someone with a great deal of credibility in the markets and finance, who
> also has y's r's d's and an n in his name, check out Ed Yardeni's y2k
> links on his site www.yardeni.com. He is the chief economist of nyc
> investment firm Deutsche Morgan Greunfell.
Yardeni predicts that we could see a serious recession due to the
Y2K problem. I agree that this is a real danger, as I posted before.
My main beef is with the apocalyptics like North who see it as the end
of Western civilization.
> Read the links at garynorth.com regarding the IRS, and you'll see that
> they're not even in the ballpark for having any kind of meaningful fix
> in time.
>
> And the fed printing up lots of cash to keep the banks from having to
> close, will only work till 01/03/2000 at best.
>
> Can someone please find an example of one fortune 2000 company that is
> even claiming to be fixed. Or one government above the county level.
> Tenessee is 70-80 percent done, but they started before almost every
> state - around 1990, I think, and most are still not actually fixing
> code. How about one major utility or telco? How bout any organization
> with over 20 million lines of code to fix.
Many organizations will have problems. But just because an organization
can't certify that it is Y2K compliant doesn't mean that it is going
to fail. The first order of business is to identify mission critical
components and work on fixing them. Then you go on to less critical
items. When the millennium comes, if some things fail you may have to
do some workarounds, but things can keep going.
> The electric utility industry looks to be screwed. Check out what Rick
> Cowles, the head of Digital's y2k utilities section, has to say at
> www.euy2k.com
Cowles is selling a Y2K book from his site. He is making money off the
problem. That gives him a potential bias. Read the link I provided
above for information about how the power grid is designed to keep
working in the face of failures. Also see the posting here by someone
knowledgable about nuclear plants debunking the fear of Y2K errors
putting them off line.
> And let's not forget the y2k errors found in nuclear weapons, admitted
> to recently by Clinton's newly appointed y2k czar, John Koskinen. He
> said something to the effect that while there is a small chance that
> they may actually accidentaly fire, they are more likely to fail safe.
> But even that destabilizes the balance of terror. Now, the world's
> nuclear powers have to wonder whether theirs and others' missiles now
> have a "best if used by date" on them of 01/01/2000.
I'm not a nuclear expert, but I believe these weapons are loaded with
fail safes and are not in a position to go off accidentally.
> About resetting clocks to 1972, that is a clever fix for a very small
> number of systems. Most embeddeds have no way for the user to change the
> date, and dos/windows machines can't handle dates before 1980.
>
> Interesting assertion: One third of business software source code is
> missing
>
> Interesting assertion 2: The embedded situation may be twice the size of
> the mainframe y2k problem, but awareness and planning is about where it
> was for mainframes 2 years ago.
It's relatively easy to identify embedded systems that could have date
vulnerabilities. They generally require some mechanism to reset the
date in the event that power is lost (since any clock requires power).
So there is little chance that embedded systems will surprise people by
turning out unexpectedly to have Y2K problems.
There are date sensitive embedded systems in some specific applications,
and testing and fixing those is going to be a big job. But I've seen
simplistic figures suggesting that 5% of all embedded systems have Y2K
vulnerabilities, and there doesn't seem to be much evidence for this.
Not much is known yet about the scope of the embedded system problem,
but more information is coming out this year as testing continues.
Hal