y2k stuff

From: Philip.Belben_at_PowerTech.co.uk <(Philip.Belben_at_PowerTech.co.uk)>
Date: Wed Jan 6 07:28:09 1999

David Hansen wrote:


> Ok, so it sounds like the general consensus on this list doesn't
> perceive any possible problems from systems that perform automatic
> disconnection/refusal of services due to systems that are based on
> time/date ranges? Here are some examples of situations that are believed
> to be possible...


No, it is not that we do not perceive any possible problems. I at least
believe that there will be problems. But I don't believe there will be
major ones. And I certainly don't believe that they will happen all
suddenly on 1/1/00.

> utilities (power/water/telephone):
> Say a disconnect period is 2 consecutive months of non-payment. A
> customer pays his/her bill regularly but if the date isn't interpreted
> properly the next time the system does a check for 'disconnect
> candidates' it will not be able to find a single payment from a single
> customer within a 2 consecutive months time period prior to 01/01/00
> (1900).


Possible. And may affect telephones. But water, gas and electricity
suppliers won't have the staff to go out and cut everyone off, since this
has to be done physically at the premises. Besides, such a huge number of
disconnections will be noticed and the cause identified.

> banks:
> Keeping in mind that banks routinely deactivate and -absorb- every
> account that is over x years idle (commonly 2 yrs. but every one I have
> come across has some variant of this), see how the above method applies
> to this situation.


Are you sure this is legal? A couple of years ago the National Savings
Bank (as it was in the UK) were saying that they had accounts that had been
inactive for over eighty (yes, 80) years with a few shillings in but they
couldn't close them because they couldn't trace the owners or their heirs.
(BTW Nat Savings never paid interest on balances less than a pound, so a
few shillings 80 years ago is still a few * 5p today)

AFAIK there is nothing in any contract I take out with a bank that says
that if I deposit money and leave it there for over 2 years they are
entitled to it. I come back 10 or 20 years later expecting to retrieve my
savings. Yes, I did this with my German bank account in 1995 having not
touched it since (I think) 1985. It was still there and I could still draw
money out when on holiday in Germany.

Yes, empty accounts may well vanish, and rightly so!

> vendors of perishables:
> Shipments of perishable items (food/medicine) are refused by automated
> systems that read the dates on the items as expired.


This has _already happened_ and will be addressed long before 2000. IIRC a
chain of supermarkets in the UK was asked by a tinned (canned) food
supplier why they were suddenly ordering three times the usual amount of
tomatoes. It turned out that the warehouse was accepting tinned tomatoes
each day; overnight the stock check found that they were out of date ("best
before end feb 00" or something) and ditched them, and the following day
more were ordered.

It is not something that suddenly happens in 2000.

> payroll:
> Employee doesn't have any hours during the 'new' pay period so no
> paycheck is issued. Also paychecks are issued with wrong dates and such.
> Clients aren't billed if there isn't anything in the billing period.
> etc...


May happen. But the first manifestations of the bug will have turned up
already - temporary contracts that expire in 2 or 3 years for example. Our
accounting, job control etc. system uses 2 digit dates. Last year (or the
year before - I forget) a lot of work was done on our system and a 2 digit
date is now assumed to lie between 1950 and 2049.

> security access:
> I'll use my company as an example. I have full access to the office
> building between the hours of 7am - 7pm M-F by way of a keycard. If the
> wrong year is being calculated then M-F can easily be Sat-Wed., etc...


As Tony and others have pointed out, this will probably occur. But it is
relatively easy to override manually until sorted out.

> credit cards:
> Accounts are deactived or non-existent.


Credit card companies - I forget who told this one from personal experience
- hit the bug a year ago with cards that _expired_ in 00. This made them
do all the necessary work then. While I concede that it is possible that
there may be a few bugs left in (spurious late payment penalties for
example), I doubt that accounts will disappear (and who will care? I'm not
going to fuss if a lot of money I owe is forgotten about ;-) ). I know
this is another "assumption" - that having hit the one bug they did all the
work to cure the others - but I think the evidence points that way.

Damn! This discussion is way off topic and I wasn't going to join in.

Oh well, I've typed it now.

Philip.
Received on Wed Jan 06 1999 - 07:28:09 GMT

This archive was generated by hypermail 2.3.0 : Fri Oct 10 2014 - 23:32:03 BST