I have written a paper, "Fear and Loathing in Mid '99?", which addresses fears that an organization's batch processing applications will not be ready for the Y2K deadline which is now less than 10 months away. I have added this paper to my Web site, which is: http://www.crosslink.net/~erington

Communications about this paper, or anything else on my site, can be e-mailed to: petere@ricochet.net

I am not trying to sell anything.

-- Peter Errington (petere@ricochet.net), March 08, 1999

Answers

Peter; Thank you for the insights. Keep well. Regards,

-- Watchful (seethesea@msn.com), March 08, 1999.

http://www.crosslink.net/~e rington/

Hi Peter. Read your bio, sounds impressive. Didn't understand much of what you wrote in the paper though ;-)

-- Chris (catsy@pond.com), March 08, 1999.

Peter: there are no silver bullets for y2k and your approach is certainly not novel. The bigger problem is that management underestimated the level of effort required for a formal fix, which was quite predictable, involving system replacement, date expansion and confusingly incompatible windowing and encapsulation techniques. What you now see is programming shops indulging in a lot of frenzied kludge coding (which is what you describe) to try and salvage a workable system.

Got mountain property?

-- a (a@a.a), March 08, 1999.

Responses to thread messages:

To Watchful and Chris, thanks for the kind words.

To a: You say my approach is not novel. I'm well aware of that. (I read a piece by Don Estes dated, if memory serves, Jan 97, referring to "firewalls" where year values were changed to leave data and programs encapsulated. This is the purpose of the temporary datasets in my proposal.)

I agree with you that there are no Y2K silver bullets that will make remediation really easy. My proposal, for example, would require considerable testing, just a great deal less than other approaches. For this reason, the longer paper on my Web site, explaining my approach in detail, has the title "Sort of a Silver Bullet". But still, some approaches are much better than others. (If time shifting had to be ruled out, I'd go to windowing.)

I have to differ with you when you label my approach a kludge. I think it is elegant, because of its economy. Avoiding messing around with the innards of application programs, that is elegance to me. But who cares about labels. My approach, I believe, will save great amounts of effort and money, and could save a lot of people's bacon, so you can call it anything you want.

-- Peter Errington (petere@ricochet.net), March 09, 1999.