Read on Twitter
Looking through Wikipedia's Y2K article it's still disappointing (though understandable) that first hand accounts like mine are flagged as unreliable, but pig-ignorant waffling by non-tech journalists are reliable, because they're a published source.
https://en.wikipedia.org/wiki/Year_2000_problem 1/8
https://en.wikipedia.org/wiki/Year_2000_problem 1/8
This, from the New Scientist, is hardly pig-ignorant. It discusses the tech issue well, but it's woefully out of touch over the management & risk issues we had to juggle. To call windowing lazy & "the worst of all possible solutions" is plain ignorant. 2/8 https://www.newscientist.com/article/2229238-a-lazy-fix-20-years-ago-means-the-y2k-bug-is-taking-down-computers-now/
There was insufficient time to do all the work we wanted. We had to take a pragmatic, risk-based approach & do only what was necessary before we hit our critical dates (which weren't going to shift). Re-writing systems, instead of windowing, would have meant that critical...3/8
...systems wouldn't have been ready in time. Windowing bought us time. We were clear it was a temporary fix, but buying decades when there's only a couple of years to save all the business critical systems was a smart move. It was up to the C level to ensure the time... 4/8
...we bought was used wisely. The NS article misunderstands the difficulty in "rewriting code". Switching to 8 digit dates (ddmmyyyy) would have entailed changing not just code, but data shared by multiple systems. That would have increased the workload unacceptably & also... 5/8
...stopped us segregating applications to fix/test/release in a phased manner. Config & release management would have grown from a complex problem to an impossible nightmare. If you don't understand release/config management issues you don't know what you're talking about. 6/8
Every article I've read that criticises how we handled Y2K makes one of 2 errors. Either it says;
1- we should have done what we did do (the author having failed to understand what happened), eg "they should have taken a risk-based approach, not fixed everything", or... 7/8
1- we should have done what we did do (the author having failed to understand what happened), eg "they should have taken a risk-based approach, not fixed everything", or... 7/8
2- we should have taken some over-simplistic alternative approach, which we considered & rejected.
We did what we did because we were experts. Deep expertise is vital. You need experts to handle scarily complex problems. Amateurish posturing doesn't cut it in the real world. 8/8
We did what we did because we were experts. Deep expertise is vital. You need experts to handle scarily complex problems. Amateurish posturing doesn't cut it in the real world. 8/8
PS. This was my blog account of my experience as a Y2K test manager. It's interesting looking back to see I didn't even mention windowing in it. That's because it didn't seem particularly interesting. It was an obvious technical response. https://clarotesting.wordpress.com/2015/01/12/y2k-why-i-know-it-was-a-real-problem/
