SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : 2000 Date-Change Problem: Scam, Hype, Hoax, Fraud

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: David Eddy who wrote (989)12/14/1998 2:20:00 PM
From: Contra Guy  Read Replies (3) of 1361
 
David:

>From one tiny angle, yes, dealing with software boundary
>issues like Y2K are "typical." What's definitely NOT
>typical is the size, scope and DEADLINE for Y2K.

Although this may seem intuitive to you, this is simply not true and in fact represents one of the biggest misunderstandings about Y2K. This myth is closely related to another that is also repeated again and again and again: many software development projects are late and cost a lot more than anticipated. That's absolutely true, and I have the scars to prove it. The point is that Y2K does NOT fall into this category.

In my experience, every single disaster that I have run into has been caused by the same flaw - a failure by the people building the system to understand what the user really wanted. To be fair to systems analysts, users themselves often don't understand what they want, or are unable to express it cogently. Or often the users are on track, but the institution's administration either overrules them, or doesn't listen. These project nightmares occur because rather than a straight path from A to B, they follow a maze of successive approximations, slowed down by a lot of frustration, finger pointing and arguments.

What makes Y2K such a breeze for guys like me to work on is the "specs" are absolutely crystal clear. Just make the thing work like it does now into the next century. No endless meetings, prototypes, alpha, beta, etc, etc, that drives all of us programmers crazy. I just finished a Y2K compliance job on a ancient legacy accounts receivable package for a group of large hospitals in Ontario. A piece of cake. Worked out to about 10% analysis, 2% coding, 88% testing. The interesting part is, I happened to be on the team that implemented Canada's new General Sales Tax (GST) about 10 years on the same piece of software. Let me tell you, that project was a nightmare.

We were working through the draft legislation trying to figure out what was taxable with a bunch of accountants. The rules were Byzantine. A package of 100 tongue depressors was taxable, but a medical kit which included a tongue depressor was not. A couple of donuts at a coffee shop was taxable, but a dozen at Safeway was not. We worked like dogs for months to get it all in there.

Now compare the implementation of the GST to Y2K for your criteria. I'm not sure exactly what you mean by "size", but if you mean amount of program coding, GST was larger by at least 2-3 orders of magnitude. "Scope" would be roughly comparable - Y2K is very prevalent, as you say, but as an MBA, I'm sure you can imagine the scope of this new tax. Every single commercial transaction in the country was effected! As for deadline, believe me, governments can be every bit as implacable as calendars. Although it might have been possible to theoretically delay the implementation of the tax, in realistic terms it was quite impossible. The government of the day backed up the deadline with severe penalties for non-compliance - there really was no choice if you wanted to stay in business. Not only did it take several times as long as my recent Y2K work, but the breakdown of effort was completely different - something like 30% analysis, 50% coding, 20% testing.

My favourite metaphor is that Y2K is like one of those lakes out in the flats of the desert after a big rainstorm. It's a rare event, and the water covers a huge area - but it is only a few inches deep. We are evaporating that water like a hot desert wind. By the time January 2000 comes along, it will be just a big mud flat with a few puddles here and there.

The lamentable thing is that the kooks and amateurs have got the public so riled up that they threaten to be a major distraction from the task at hand. A colleague was telling me about a professor who was trying to find out if his wife's hairdryer was Y2K compliant. It's a hilarious story, but unfortunately, these ridiculous notions waste the time of competent people. So that's why I'm leaning towards putting a bit of a lid on things. It's better to concentrate on solutions and just getting on with the job so that people can enjoy a nice New Year's party rather than hunkering down in some bunker by candlelight.
Thanks for your comments.

Ben
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext