Telegraph | News | Wrong problem, wrong solution: "Sci-fi panics such as climate change are dangerous because they distract politicians from what really needs doing. Y2K bug: correct solution, laugh; actual solution, Y2K Office. Result: nothing, at great cost. Energy shortages and climate change (if you believe that man is responsible): correct solution, go nuclear and reverse 20th-century deforestation. Actual solution: windmills, rampant deforestation, EU paying farmers not to plant trees or anything else. Result, energy crisis, species loss and no fall in CO2."
Sunday, November 12, 2006
Telegraph | News | Wrong problem, wrong solution
Subscribe to:
Post Comments (Atom)
6 comments:
This is quite an article! Thank you very much.
The preceding article by Christopher Monckton is, if anything, even more illuminating, since it goes into the science of climate change.
Part of his intro:
Sir Nicholas Stern's report on the economics of climate change, which was published last week, says that the debate is over. It isn't. There are more greenhouse gases in the air than there were, so the world should warm a bit, but that's as far as the "consensus" goes. After the recent hysteria, you may not find the truth easy to believe. So you can find all my references and detailed calculations here.
His references and calculations are there as well.
Cheers,
tt
Disagree on the Y2K uselessness. I was just chatting with a fellow last week who spent a great deal of time fixing Y2K systems. He proudly told me that "Only one of my systems broke after the changeover." Clearly, there was some danger there. Was it the case that a very serious problem was mitigated after a huge amount of work and so nobody saw it, or was it the case that there was no problem at all? Probably somewhere in between.
...or was it the case that there was no problem at all?
Well, there was the abysmal idiocy of having the potential problem in the first place. Anything that stupid is too ugly to live, so I'm glad it got fixed.
When I needed a portable timestamp for data that might be around for a while I used eight bytes for the Julian date in milliseconds; the astronomers had that problem solved long ago. Sure, eight bytes is a bit extravagent, but even on an old 4k pdp8 it wouldn't have been a problem, and I was using a Vax730 at the time.
chuck
All well and good to complain about lack of foresight.
The programming I did was not in the scientific field--it was for business. Accounting, inventory, sales, customer info, stuff like that.
The pressure was always time and we RARELY had the opportunity to do anything from scratch. Pick up an old system and modify for a new customer was pretty much the extent of it. And since we billed by the hour, the clients were not willing to pay for changing the date field--especially when they knew they could worry about it later, like closer to 1999.
And that's exactly what they did.
And it worked out fine.
Sometime before the 2004 election someone challenged me regarding Y2K who had also claimed that I'd never explained why I was voting for Kerry and I responded that he obviously had missed the comment where I wrote about what I was doing in 1999. Roger's archives are finally being cleared so those of us who were commenting there before July 2004 (as of today anyway) can no longer retrieve what we or others of our acquaintance once wrote there. [If you follow the threads here you might be able to guess why I found this out.]
In 1999, a great deal of my time was devoted to making all of the many, many machines in a cable television advertising sales division "Y2K compliant" because that is what we'd been directed to do and because AT&T was in the process of buying the cable provider who owned us during that year and AT&T would not complete the purchase until their people were satisfied that we were "Y2K compliant." By 11:59 PM on 12/31/99, AT&T was out of the picture and Comcast was going to purchase the operator and the part of the company for whose computer systems I was responsible for well over two billion dollars. They'd offered somewhere under 800 million for the same a couple of years earlier. At 11:59 PM on 12/31/99 eight to ten of us, including my wife, were standing around a buffet on a filing cabinet outside the control room in the building where most of these machines were housed talking and eating and waiting for nothing of note to happen. Nothing of note commenced to happen soon thereafter and our little party broke up and went home. It wasn't until a few days later that someone (almost undoubtedly from the security company that'd been hired to protect, among other things, the machines) stole a working laptop and we came to find that he or she had stolen four non-working ones prior to that theft.
Back to Y2K. One effect, and I'd argue a major one, of Y2K is that people who would have otherwise waited another year or two or three bought new systems in 1999 and then didn't buy in the out years and a number of professionals (and non-professionals) made a lot of extra money in 1999 and were out looking for jobs in the out years.
Years before I'd been a programmer and we'd dealt with the Y2K problem by starting with a date specific and timing everything out in date calculations for the next 32,767 days (I do miss binary calculations) which took us well beyond 2030. None of us seriously thought back then that the programs we wrote in 1985 would still be in use in 2000, but we did allow for that possibility.
I never took Y2K seriously and I never advised anyone in 1999 to buy a new computer or router, but I did take my pay for keeping the machine running and the machine did run.
Post a Comment