Eine Kleine Nichtmusik

Witty and pertinent observations on matters of great significance OR Incoherent jottings on total irrelevancies OR Something else altogether OR All of the above

Tuesday, October 30, 2007

Oh for the love of Hephaestos....

In his Saturday column in the Guardian, Simon Hoggart mentions a book by Christopher Booker and Richard North, Scared To Death, which discusses various media-hyped scare stories including "the most perfect scare of all, the Y2K or millennium bug terror, which was going to end civilisation.......In the end, among other minuscule catastrophes, some bus ticket machines in Australia failed to work."

We in the IT profession have become used to taking the blame when highly-visible SNAFUs occur, and to being taken for granted when our efforts mean they don't. With Y2K, though, we weren't so much taken for granted as accused of having manufactured the while thing as a way of boosting our earnings. Ed Yourdon (a demigod to coders of a certain age!) made some pertinent observations right back on 1/1/2000. But I think the best rebuttal of the Hoggarts is to be had by looking at the birds and the B(MW)s. Let me explain.

Back in 1972, the environmental movement was just gathering momentum. Edward Goldsmith came and gave a talk at my school in which he discussed the Club of Rome's report Limits To Growth, on which his magazine The Ecologist had just run a feature. This was the famous report that informed us all that the world's oil would run out in 1992. (As you can see from the link, it wasn't quite that simple in fact, but the environmentalists themselves were happy to be misinterpreted in that way as it garnered more attention). Ten years earlier, Rachel Carson's book Silent Spring had alerted the world to the devastating environmental effects of pesticides, and painted a bleak picture of a future where many species such as peregrine falcons would be wiped out.

Well, as you can see, it didn't happen. There's still plenty of oil and plenty of peregrines. So where did it all go right? It went right precisely because of Rachel Carson and the Club of Rome. People read Silent Spring and suddenly became much more aware of the toxic stuff (to humans too in many cases) they were scattering around their houses and onto their crops. So they stopped using them. Not overnight, and not totally, but to a considerable extent. Limits To Growth told the oil companies nothing they didn't already know, but suddenly their investors were more easily persuaded to put money into prospecting in places where extracting the oil might be expensive, and into developing fancy techniques to permit such extraction. (A friend of mine wrote a text-book back in the 1980s on drilling round corners: in 1970 there would have been little market for it.) As the price of oil rose thanks to OPEC's machinations, it became steadily more economical to extract oil from unlikely sources, and new supplies kept on coming onstream.

So both Silent Spring and Limits To Growth were wrong in the end, and were wrong because they had first been right, and been taken seriously. If they'd been dismissed as more cranks (as they were by some folk) then we'd be living in a different world today. In the same way, if nobody had taken the Y2K bug seriously there would have been some rather worse things happening on 1st January 2000 (or maybe 4th January, or 29 February) than bus ticket machines failing. In Bank of Scotland, for example, the first sign of the approaching millennium was in (I think) 1998 when the first batch of Visa cards with an expiry date of 1900 was produced. That was easy to fix (give all new cards a two-year rather than three-year life) but provided a concrete example of the kind of thing we were working to prevent. Over the life of the Y2K project (on which I was one of the lead testers for the bank's batch systems) we uncovered a few instances of actual Y2K bugs, none of them dramatic, but by the time 2000 rolled around we could be pretty sure that we weren't going to encounter serious problems. And when I say serious, remember that the entire banking system, world-wide, depends on the electronic transmission of money; that these transactions all have date and time stamps; and that a transaction coming in with the wrong date will usually be rejected. Or imagine that the program calculating how much interest you owed on your mortgage decided that between 31st December 1999 and 4th January 2000 you'd accrued 36,528 days' worth instead of four.

So poke fun all you like, Messrs Booker and North. I'm happy to have done my job so well you never noticed.

Oh, and don't all post comments telling me that the year 2000 wasn't the millennium, because I'm well aware of that. That's why I call it the Y2K bug, not the millennium bug.

5 Comments:

At 30 October, 2007 08:28, Anonymous Phil said...

So poke fun all you like, Messrs Booker and North. I'm happy to have done my job so well you never noticed.

Damn right.

I wonder if anyone's collecting examples of surviving Y2K bugs? Earlier this year I came across a Web form which insisted on entering the year as 19107 - which, of course, failed validation.

 
At 31 October, 2007 19:49, Blogger King of Scurf said...

And of course we all earned a fortune out of it didn't we. Well didn't we? I did a 12 hour day on 1/1/2000 from 7am. I got Saturday pay. Nothing happened apart from we all got a bit skittish on sugar because the only thing to eat was doughnuts as our employers had insisted we work but had decided not to ask any catering staff to do the same.

 
At 01 November, 2007 03:17, Blogger Rob said...

I had to set up a test where we ran as much of the work of the first business day of 2000 as was available at 31/12/1999, in the window of genuine two-thousandness before the first working day of 2000 actually rolled round. We dry-ran it over the August 1999 Bank Holiday weekend and found various operational glitches which were duly fixed before the real thing. I didn't make a fortune out of it (I've worked on more lucrative projects) but I did get overtime for both the August BH and for the millennium period. The seriously lucrative overtime rates kicked in IIRC for six hours each side of the Y2K midnight, and I did in fact have to go into work for about an hour and a half late on Hognanay to download some files. That was nice. Because of what I needed to do with those files, I had to be accompanied by a female colleague from another department who is best described in the immortal words of Duckman: "Hummana Hummana Howwa !" I have to say, as I sat there basking in her glow and downloading files at some unearthly hourly rate, I did think life was quite good.....

Mostly, though, it was a normal weekend overtime rate. And no, we didn't get any catering provided (though I've done other weekend mega-shifts for projects where quite decent facilities were made available).

 
At 01 November, 2007 03:19, Blogger Rob said...

Amd Phil - a collection of surviving bugs would be great! Yours is the first one I've heard of, though I'd often wondered if there were survivors.

 
At 01 November, 2007 13:03, Blogger Chip said...

I worked for Netscape at the time. In April 1999 they cancelled all end of year holidays. Since I'd already booked my time to be here in Edinburgh I offered my resignation.

Instead they offered me a chance to work remote (sweeeeet). When it came to be Dec 31, 1999 - 11pm (US version of the date...) I was standing on Princes St waiting for the fireworks. At 6am the following morning, I was logging into my computer to see if all was well with the world - and answer questions if there were any. There weren't any... and so I took my fairly drunk body to bed.

The time between April and December had been spent creating "fixes" to avert any issues that might arise on Jan 1, 2000 - so when it came, there were no issues left.

Unfortunately, my co-workers were required to be at their desks rather than taking in the parties in California. Through some grace, I was allowed my holiday (and actually paid for it - as I was "technically" working). What most of the management doesn't know is that the workers knew the problem wasn't a problem - but management didn't want to listen to us.

 

Post a Comment

<< Home