It’s hard to believe that this time 20 years ago, the world was preparing for the new millennium. From the infamous “Y2K Bug” to stocking up on canned goods, the months leading up to Y2K was a time of bracing for the unknown. For Seattle City Light, preparing for Y2K took on many forms. While there were minimal issues, some employees did incur a few tense minutes leading up to and following midnight, Jan. 1, 2000.
This is an oral history told from the perspective of five City Light employees of how the utility prepared for the event and where they were the night of Dec. 31, 1999.
1999: A Year of Preparation
Steve O’Kelley (Communications, now retired): The way it was portrayed on the national scene was that planes could fall out of the sky and power would go out everywhere. There was a lot of hype to it. I say the average person I worked with didn’t believe that any of that could happen…probably.
Peter DiTuri (Sr. Management Systems Analyst): Up until that time, a lot of computer systems didn’t have the memory or disk space like the smartphone you have in your pocket today. When programs were written, they took up as little space as possible. Memory and disk space were expensive, so to get around this problem, programmers made applications that only used two-digit years like ‘89, ‘67, ‘04, etc. That works fine until you had to include something starting on Jan. 1, 2000 or later. Then, you had to store the full date and even calculate differences between dates, and not making 2-digit years into 4-digit years resulted in wrong or insensible values. Although Y2K was anticipated for many, many years, procrastination turned Y2K into a crisis.
“There is a lot of information available on Y2K, some of it accurate and well-reasoned; some that is misleading and unduly frightening. The City of Seattle is determined to clearly communicate the City’s progress towards readiness and to provide you with the resources you need.”
Bill Davis (Telecommunications Supervisor, now retired): We didn’t have computers the last time we went into a new century. So no one really knew for sure what was going to happen.
Peter DiTuri: The City of Seattle was on top of this. The Executive Services Division took the lead for Y2K remediation efforts. Each department had to identify each individual application or system hardware or software that had a danger of failing if not remediated or replaced by Jan. 1, 2000. They could be small databases on someone’s work computer, giant enterprise applications or major pieces of equipment. Y2K was a 100% human-caused problem but with a 100% known solution.
Back then, we had an IBM mainframe. The problem with it was that IBM no longer produced the parts for that machine, and the operating system wasn’t being updated, so it had to go away. City Light had its Transformer and Network Load Management applications on the mainframe and we had to figure out what the lost source code was for the applications, then we (this is really cool) disassembled the application, wrote flowcharts of pseudo-code, rewrote the application code in Fortran, correcting for Y2K (of course) and imported it over to the Sun Solaris server where it ran for many, many years after. So getting a program off the mainframe was a necessary solution for Y2K as well as updating lots of source code and changing programs to handle four-digit years.
Bill Greif (Sr. Data Processing Systems Analyst): The team came up with a plan for how to check everything that was running and test it out. We went through a lot of tests. We checked the equipment in a test environment to see what would happen when the clock inside the equipment went over into 2000.
Peter DiTuri: Other programs we had to update so they would accepted dates past Y2K up to 9999, we won’t be here for that occurrence. But we did have to test for other dates, too. We had to make sure that Feb. 29, 2000 was also valid since 2000 was indeed a Leap Year. 1900 was not and neither was 2100. Y2K was a continuous process to make sure dates like that still worked.
Steve O’Kelley: One of the things we did as a group was to identify the equipment that could be susceptible to a problem from Y2K. A contractor inventoried our equipment. I worked on telephone switching equipment and some communications gear that switched out to the substations. The feeling in the group was that nothing much was really going to happen.
“Four years of hard work and preparation will be put to the test Friday evening, through the following Monday, the first business morning of the new millennium. I am confident we’ll ace that test.”
-Gary Zarker,
City Light Superintendent, in a memo on Dec. 27, 1999
5…4…3…2…1…Darkness
Peter DiTuri: We had the plan for staffing for the night of in case something went wrong. There were dedicated workstations and PCs in our Trouble Center (Seattle Municipal Tower room 3613) in case there were any major outages or any problems major to our system. We did have one workstation fail after zero hour, but we isolated the issue to a machine configuration issue and I was able to fix it fairly quickly.
On that particular day, (Dec. 31, 1999) I was working in City Light IT, our “war room” was in SMT 2921, and we had a pizza party. We also had mandatory overtime. Management and the unions had agreed that this was an odd situation, so we in IT staff assigned to staff Y2K were set to work up to 24 hours – but we never worked that long. I worked from 6 p.m. to about 3:30 a.m. New Year’s morning.
As we came closer to zero hour, we watched the Space Needle on the TV and we knew it was close to time. I had my CD player so I took it out and I popped in a CD and it started playing REM’s ‘It’s the End of the World (As We Know It)’ [laughter] I could not take Y2K seriously. At that point, we had reports from across the world and nothing was going wrong! It was like…why are we even here now? But you never know what could have happened if something crucial didn’t get fixed as it should have.
Steve O’Kelley: I was at the System Control Center (SCC) at midnight. I didn’t pick up any great worries. We turned the conference rooms into media centers so that the TV stations could plug in if they showed up. We had TVs set up so they could pick up KING and KOMO.
Bill Davis: Everything went fine until around midnight when the lights went out at Queen Anne. Everybody thought at the time ‘well oh shoot, something went wrong.’
“SEATTLE – Just a few moments after midnight there were a few power outages in Seattle. But at this point it’s unclear of this was Y2K computer problems or something else. Seattle City Light reports about 1,500 customers were without power momentarily. All power was restored.”
-KOMO
Bill Davis: It turned out to be mylar balloons caught up in transmission lines around midnight. Everybody breathed a sigh of relief. We had all the media down at the SCC. The City Light superintendent was down there so everyone was there watching. I went to each station to hand out cookies and non-alcoholic champagne and tried to stay out of the way of the dignitaries and the press. Then they let us go home around 2 a.m.
Bill Greif: NYE was actually a pretty fun night. We had the news media here [at the SCC] and we were all hanging out in the control room waiting for something to break. One of our employees had a yellow ergonomic ball chair and put a smiley face on it and put it on the ladder we use to get up to the pinboard here. He tried to set it up for it to roll down at midnight but it started early and made the news for that.
Alone on Y2K
While City Light employees prepared to welcome in the new millennium across Seattle, Rusty Welshon anxiously waited with some concern at City Light’s remote Tolt Powerhouse.
Rusty Welshon (Electrician Constructor and Operator at Cedar Falls and Tolt powerhouses): I was at Tolt which is all computer-controlled. It uses Programmable Logic Controllers (PLC), an industrial process computer. Tolt was the first real computer-controlled powerplant that we had. All the other ones had a hydraulic, mechanical controlled system that was more manual. We had a PLC and an interface between you and the machine called a Human Machine Interface (HMI), which had a series of screens that you could rotate through. Two months prior, we did a trial run by changing the date of the PLC to Dec. 31 at 11 p.m. and let it run to see what would happen. The HMI screen twisted out of shape and the numbers were all distorted.
We suspected that…We didn’t….ummm…(chuckle) we didn’t know what the heck it was what going to do.
We couldn’t get any information out of it, and we couldn’t get into the program. We were concerned, not overly, but we were concerned because we didn’t know what it was going to do.
Tolt’s a smaller plant, only 16MW so it wasn’t critical for it to be in operation at midnight on Dec. 31. We turned off the generator because we didn’t know what was going to happen.
So 11:59 p.m. rolls around, I’m all by myself in the middle of the mountains up above Carnation. I have a flashlight and the PLC ready to be plugged back in and I have my hand on the hydraulic handle so I can shut down and freeze everything in case things went catawampus. If the hydraulics started to move, things could have gotten scary so I was ready to lock it down if needed.
And sure enough, the PLC distorted almost as if it was a crumpled piece of paper. Then it displayed weird numbers. So what had happened during the trial happened again. So I plugged my computer back in and waited until 12:30 a.m. so whatever would have happened will have happened. I fixed all the numbers and called the dispatcher to see if I was good to get it running again. I started it up after adjusting the numbers and everything was found. It was anticlimactic, but it was not Y2K compliant so there were wrinkles and we had to adjust. And we didn’t know what was going to happen. To a degree, it was a tense moment because Y2K was a real thing and it came upon everyone faster than we thought.
Surviving Y2K and Moving Forward
Peter DiTuri: Planning for Y2K was a team effort. It really was. Everyone was on board and knew what and what the City needed to have in place. If anything, Y2K made us define our business processes as a utility and what our points of failure could be. We had to outline our processes and figure out alternatives in case something bad were to happen. It still applies today. We are now more susceptible to failures than in Y2K because of our reliance on the Internet, and many mistakenly assume that disruptions from disasters only result in short-term, minor inconveniences.
Bill Greif: We were ready to go into manual mode if necessary. We were pretty confident that if something did happen then we would be able to manage it. Other than that, we had done so many exercises prior to make it a non-event. I was on the phone with a friend around midnight who was kind of a prepper who said ‘just wait, Bill’ and then midnight hit and nothing happened. And then he said ‘yeah, it’s still early, Bill. Just wait. Shipping will quit and we won’t be able to get food.’ which never happened.
Bill Davis: It was kind of a good experience, having all of those units working together and help each other out. It was an interesting time.
Steve O’Kelley: It did get kind of hyped. The internet was kind of around but not a lot of people had it. So you believed what was in the paper and talk radio. My biggest regret was not keeping this one cannister of hot cocoa that had an expiration date of 1901, not 2001. That’s where they made a subtraction of the dates. Those are the kinds of things that got most people, using the last two digits.
When I saw a request for stories about Y2K, I thought to myself ‘wow, I can’t believe it’s been 20 years.’ It seems like yesterday, but it wasn’t.
A special thank you to those who contributed to this oral history.
If you were a City Light employee during Y2K, we would love to hear your story! Email nathan.macdonald@seattle.gov with the subject line: “Y2K at City Light.”