In the late 1990s, humanity became obsessed with the impending Millennium bug. Until the year 2000 came and the lights stayed on, planes did not fall out of the sky, and life went on as usual. Where did this fear come from? And was it justified?

In 1999 Jean Popma celebrated New Year’s Eve without his family and friends. Work came first. And his was quite a task: as Vice-Director of the University Centre for Information Services (UCI), he was appointed head of the command centre – located on the Geert Grooteplein – responsible for smoothly navigating the campus through the turn of the millennium.

As the clock neared twelve, the tension mounted for Popma and his team. ‘We’d worked towards this moment for two years. Had we forgotten anything? We thought we had it all covered, but then there are external factors. What if there was a power cut, for instance? Not everything was under our control, and we had to anticipate every single scenario.’ The command centre was equipped with emergency lighting and computers running on battery power. Just in case.

Just in case what? A few years before the turn of the millennium, an increasing number of experts had begun to express their concern. They predicted a problem in software systems: these systems dated from a time when data storage was limited, so in the 1960s programmers had built systems that indicated years using two digits instead of four. The year 1995 was abbreviated to 95, 1996 to 96, etc. But what would happen if 2000 dawned, and computers were sent back to the year 00?

There was much philosophising on the issues that might arise. Some devised the most bizarre scenarios, with planes crashing down, and Russian nuclear missiles firing at will. Banks would no longer be able to issue money, and food supplies would be exhausted in no time. Not everyone remained in the Netherlands, waiting for disaster to strike. Take Frank Slobbe: this computer expert and ‘millennium refugee’ was frequently featured in the media in 1999 for moving to Australia. After all, they didn’t have nuclear plants there, he reasoned (only to find out upon arrival that in fact they did – Oops!).

On the Nijmegen campus, people were also anticipating all sorts of scenarios. After all, a large institution such as Radboud University is dependent on all kinds of software. Would salaries still be paid? Would the doors of the buildings still open? The hospital had to be ready for any contingency; patient care comes first.

Back to the command centre on the Geert Grooteplein, and that fateful New Year’s Eve. Popma and his colleagues sighed in relief as midnight came and went, and all the lights stayed on. They didn’t need the extra set of warm clothing the ICT experts had brought along with them: the heating was still working.

While fireworks lit up the city, Popma and his colleagues opened a bottle of champagne. They had a quick drink and went back to work. In anticipation of Millennium night, many systems had been preventively switched off. These all now had to be restarted manually. The mood was not quite euphoric, but Popma’s team was definitely relieved. At 4 a.m., his shift finished and Popma went home.

Hilarity
The mild sense of panic that had preceded New Year’s Eve quickly turned to hilarity about the Millennium bug. Was this what we were all so afraid of? Places where the bug had done damage were hard to find. Newspapers wrote about a Japanese nuclear power suffering a power cut, a Norwegian hospital where the medical equipment faltered temporarily, and a French and American satellite that refused to work. And that was it.

This raises the question: Was the fear of the Millennium problem actually justified? According to Joost Visser, Professor by special appointment in Computer Science at Radboud University, despite the relative lack of problems, the answer is yes. ‘It’s easy in retrospect to say that there weren’t many problems, but that’s also because the problem was taken seriously. I’m absolutely certain that there was a problem, and people worked hard to fix it. Whether the degree of preparation was justified by the actual scope of the problem… I don’t know. All I know is that we successfully met the Millennium challenge.’

Jean Popma, who now works as project manager at the Radboud University Digital Security Department, agrees. ‘We’ll never know whether we took the problem too seriously. But if the University hadn’t done anything, we would have had lots of problems. Solving all these problems after the fact would probably have cost a lot more money than the preventive investment.’

And this was no small investment. Popma estimates that the University spent approximately two million guilders on a problem-free millennium transition. ‘This money was spent among other things on extensive software testing, including night-time time travel to change the date under controlled circumstances. We also preventively updated or replaced a lot of outdated systems.’ The Executive Board was pretty nervous, as Popma remembers – after all, the administrators were ultimately responsible. ‘We were frequently asked to report on our preparations.’ Now Popma can laugh about all the millennium bug fuss. ‘It was both an exciting and funny time.’

A time that was also very informative for lots of people. ‘As the turn of the millennium approached, people became much more aware of how closely technology was interwoven with their life’, says Popma. Airports, nuclear plants, hospitals: they all ran on computer systems. ‘People weren’t always aware of this. If all these institutions were to falter, it would disrupt the whole of society.’

Joost Visser: ‘Prior to the year 2000, ICT was primarily seen as a technical service – as if it was only about automating processes. People thought technology was simply something you needed. With the turn of the millennium people began to realise how dependent we are on computers.’

Doomsayers
Although we still have nearly one thousand years to go before the next millennium change, software problems like the millennium bug may also arise in the near future. According to Professor Visser, the millennium bug is not solved yet. “Institutions such as the central government, banks and Statistics Netherlands work with software that dates back to the 1960s and 1970s.” He explains that these systems are not built up from scratch, but are continuously expanded and adjusted. The original programmers have long since retired, so the ICT experts working on these systems now don’t know precisely how they were constructed. The millions of lines of code in these systems could still contain many errors that may surface in the future.

A specific problem is expected to arise in the year 2038. Many computers run on a UNIX operating system dating from 1970 that keeps track of time in seconds. The problem is that in 2038, this system will reach its ‘maximum’. If the structure of the software is not changed before then, on 19 January 2038, computers will think that we have gone back to 1901. What’s more, and this is where it gets exciting, to a Friday the 13th in December. This combination of circumstances is food for thought for doomsayers: is the world going to end soon after all? With the millennium turn behind us, we still have something to fear!