y2k countdown: Real Deadlines Produce Strange Behaviors

I am sure many of you have your own Year 2000 horror stories to tell, but hopefully not as bad as mine. While my story is continuing to unfold, it couldn't have started any worse.

Earlier this year, as I was switching consulting projects, I asked my boss about the prospects of working on a Year 2000 effort. Instead of a blessing, my boss told me Year 2000 remediation was not in the strategic interest of our company. She even quipped that this was a short-range problem that would force us to look for more revenue within a year. So I chose another project; why fight the boss?

Before I had a chance to start this new project my boss told me that a request had come from our biggest client to support his Year 2000 project. The customer had just been appointed Y2K Czar for his company and now our company suddenly had three consultants supporting a non-strategic interest project.

When the Czar called his first meeting, I immediately noticed something strange. Although I knew his company had a Year 2000 team in place for the last two years, there was no one else present from his company and he did not reveal any project history. I wondered what skeletons were hidden in the closet or if any work had been accomplished.

My company's mission was to do functional end-to-end testing of his organization's communications and data centers. I asked for the inventory and the functional connectivity diagrams and was handed a spreadsheet with all kinds of strange nomenclatures and "N/A" located in most of the "Y2K Lines of Code" column. When I asked the Czar how I should interpret this data, he took a close look and agreed it was meaningless. So I decided to go right to the source and asked the Czar to tell me who prepared the spreadsheet. He said "Mr. Jones." I then asked for the room number, phone number and first name of Mr. Jones. Strike three on me! The Czar did not know and could not find Mr. Jones in the company phone book. He said to check out the former Year 2000 office across the street. I prayed this was not a cemetery.

Immediately after the meeting I went to the building across the street and conducted a room-to-room search for Jones. After no luck on the first floor, I hit the second floor. The good news was they knew Jones. The bad news was they wanted to know why I was asking these probing questions. I told them I was working for the new Y2K Czar and needed an explanation on Jones' Year 2000 database. I was given Jones' room number and told it was in the Y2K office on the third floor -- I guess the Czar had not let this organization know that he was now in charge and this was the "former" Year 2000 office. I had a feeling these people were no longer on the project but they did not know it yet.

Jones turned out to be a nice enough fellow. He explained he was busy preparing Year 2000 corporate status reports, but promised me a detailed spreadsheet in a week.

A week later, Jones was too busy to talk with me because he was still bouncing off the walls from his own missed corporate deadline. He told me to see one of his minions for the data.

Mr. Minion had the spreadsheets printed out on legal paper. He mentioned that most off the data was stale due to the updates made in the last week for the corporate report. I asked him to print out the update. While the update was printing and Minion was explaining the meaning of the data, the door flew open and a suit demanded to know what was going on.

Since Minion was speechless, I explained to the suit the whole Year 2000 story. Mr. Suit told me that he was a senior VP and his reports had to get out the door. He did not want me in his building preventing valuable work from being done until the reporting period was over. So I grabbed what data I could and left.

When I reported the ambush to the Czar, he informed me I would have to work through these problems since operational reports take precedence over Year 2000 testing. Thanks for the warning.

I went back to my office to ponder a strategy with the limited information at hand. While there, my VP stopped by for a chat. He informed me that he had a strategy to test the effects of the Year 2000 on computers: Roll the date forward to the century and see if the computers worked. Stunned, I informed him that working on production machines was risky and asked if he had taken into consideration that rolling the date forward could expire his software licenses. The VP was awe struck. None of this ever occurred to him. I did not inform him that most companies have a termination clause for amateurs messing with Year 2000 testing.

So much for my first week.

What did I learn? The only difference between the Year 2000 and other software problems is that there is a real deadline this time around that causes the irrationality to be compressed.

Bob Lynch is an engineering consultant working on the East Coast. He can be reached at lynchr@acm.org.