BK: Thanks for another well considered post. I am keeping printouts of the original post ghsport.com and your response Message 8691161 as the best articles "for and against" the arguments, although both of us seem to have a "middle of the road" feeling about the problem ie. there will be problems, but not catastrophic.
You are of course correct in the Y2K problem being a design flaw, although the "Y2K bug" label has been used so often, it is now part of our logic. Would you not agree that most computer "bugs" are caused by design flaws ie. not allowing for negative numbers, missing component fields, or inadequate data entry edits?
Fortunately, the computer software we sell uses the PICK Operating system, which stores all dates as the number of days from 31Dec67. For example, to-day is 11417 and 01Jan2000 is 11689. Because I do not have to fix any software, perhaps I have under-estimated the work that must be done to fix those systems which still store the date as YYMMYY. I thank you for sharing your experience with this problem. However, I read that the average computer software is in use for seven years. I wonder how many programs written in the past 10 years do not allow for the year 2000. If they do not, the system analysts (designers) were especially negligent.
Since all our application software and the operating system does not care whether the date is 11688 or 11689, I find I have to spend all my time defending the fact that we don't have committees and action plans to "fix" all our systems and programs. In fact WE are being considered negligent even though we did it right the first time!
The only reason I got involved in this debate is to try to argue some reason to those who think that ALL services will end next January, and we will be left in the dark, in the cold, with no food or transportation. I agree with you that the preparations some people are planning out of panic might cause many of these problems ie. taking money out of the bank, stock market, hoarding food, drugs etc. The last straw is when I read a post on one of the threads questioning whether pace-makers would fail!
In your response, you state that you are most concerned about the possible failure of embedded chips. I agree that this could be a huge problem, but took comfort from the article that: Just because a device deals with time, doesn't mean it cares about the date. If it does use the actual date, it must have a date-input function for initialization after power has been removed.
This would seem to limit the number of chips or embedded systems that could be a problem, and to quickly identify those that must be tested.
Would you agree with this assumption and conclusion?
Thanks again for taking the time to consider these arguments and to give such reasoned responses. I suspect that we tend to agree on most if not all issues, and perhaps only differ slightly in the impact of the Y2K "design" problem.
Jim |
