Almost everyone will have heard by now - though few believe - about the doomsday scenarios of planes falling out of the sky, nuclear power stations going out of control and the onset of worldwide recession when the 'millennium bug' hits information systems around the world in the year 2000.
Short of not being able to get money out of a cashpoint on New Year's Day to buy that vital hangover cure, what are the real implications for geotechnical engineers when the clock strikes midnight on December 31, 1999?
There is widespread concern that while many businesses are aware of the Y2K (Year 2000) problem, they have not started doing anything or are not doing enough to combat the possible disruption caused by computer failure. This is partly due to the perception that only older systems will be affected, and also some scepticism (although this is diminishing) that the whole thing has been blown out of proportion by companies that will benefit from the extra work created.
This potentially disastrous attitude was highlighted in a report last year by the Department of Trade and Industry that estimated only 3% of small construction companies have computer systems which could cope with the millennium problem. More than half said they would wait until 1999 to start tackling the problem.
That may be too late. According to international law firm Clyde&Co, some IT companies are recommending that systems comply by the end of 1998, to allow a full year before the first of the major problems begin. 'It is a problem which businesses cannot afford to ignore and must address now,' it says.
The Y2K problem stems from the computer industry's relative infancy in the 1960s and 1970s. Memory was at a premium, both in physical and financial terms, and programmers discovered that by using only two digits for the year in date fields (implying that, for example, 78 was in fact 1978 and not 1878 or 2078), a great deal of space could be saved. No-one envisaged the systems would still be in use 30 years later and so the method became common practice.
But when '99' becomes '00', hardware and software systems with date sensitive information could run into difficulties. Because they will not know whether '01/01/00' means 1900 or 2000, data may not be recognised. The computers may think they have to reset or, at the worst, become corrupted and shut themselves down.
Problems may well snowball because 2000 is also an 'uncommon leap year'. Years divisible by 100 are not leap years, unless they can be divided by 400. In other words, some systems will not realise that 2000 is a leap year and could experience problems on either 29 February and 31 December, both 'unexpected' 366th days.
It gets worse. Some systems look forward one year to calculate dates and could easily run into difficulties on January 1, 1999, so problems will start even before 1 January, 2000. Other numbers entered in date fields are significant in programming. For instance, September 9, 1999 is symbolic in some programs because a string of nines represents the last date files need to be kept until. And '00' was also used in some systems to represent 'not applicable' or 'beginning of file', which could create further chaos.
While the Y2K problem was originally thought to be concentrated in mainframe computer systems developed in the 1960s and 1970s, it is now known that software and hardware created up until very recently has a good chance of being affected. Indeed, personal computers manufactured before 1996 are unlikely to be year 2000 compliant. A recent investigation by Prove It 2000, an independent computer auditing company, revealed that 80% of PCs on sale in the UK today failed at least one compliancy test. It is also claimed that more than two thirds of software on sale has some sort of vulnerability.
Because date sensitive information is present in almost all IT systems, the knock-on effects could be very serious, with hardware, embedded systems (eg railway signalling, security and fire systems), computer operating systems and a variety of software applications hit.
But exactly which systems will be vulnerable and how they will be affected is not clear. It is thought that financial software used for company payroll and ordering systems is likely to be affected as it relies heavily on dates. For the geotechnical industry in particular, design packages with date dependent data and instrumentation (both field and laboratory) that has embedded computer processors or dedicated hardware and software could run into trouble.
The only thing companies can do is to carry out an audit of their computer systems, either internally or by employing an independent company (and with time running out it is suggested they look at those most critical to business first). However, systems have been known to pass one check and fail another, so this is not an absolute guarantee of compliance. But until this stage is carried out, work to rectify the glitches cannot even begin.
Who should pay for this work? Taskforce 2000, set up by the Government in 1996 to raise awareness of the Y2K problem, estimates that the cost in the UK alone could be more than £50bn. Costs could be crippling to small business, which make up a large segment of the geotechnical industry. Worldwide, Taskforce 2000 estimates 1% of established businesses could go bankrupt as a direct result of the millennium bug.
It might be thought that responsibility only lies with those who wrote the software in the first place. But in reality, their liability may be difficult to establish: they may no longer exist as a trading enitity. Clive Seddon, partner in the Information and Technology Group of Masons, Solicitors, says companies that design IT systems are obliged to exercise reasonable care and skill in performing their work under a contract. Contractors can in some circumstances use the 'state of the art' defence. This means the IT company can argue that although the product is not Y2K compliant, until very recently competent programmers represented years by two figures. Therefore they exercised the necessary skill, so there is no breach of warranty. It's a line of thought that all engineering practices will be familiar with and presumably sympathetic to.
However, Seddon adds that IT companies do face substantial risks if they have expressly warranted that the system supplied is 'fit for purpose'. This means that the contractor will be liable, regardless of the level of skill, care or expertise that has been employed in the design or programming. 'If an IT system fails completely due to the Y2K problem, it will be difficult for them to maintain that it is fit for purpose,' he says.
Companies that either commissioned or wrote their own software will also probably have to take full responsibility for the necessary work. In any case, it looks as though most companies will have to meet the costs of compliance before trying to claim compensation.
However, industry should not feel that it has been left holding the Y2K baby without the prospect of outside help. There is a code of practice produced by the British Standards Institution, DISC PD2000-1, A Definition of Year 2000 Conformity Requirements. This is now the widely accepted working definition of compliance/conformity for the construction industry. It lays down four rules of compliance:
no value for current date will cause any interruption in operation;
date based functionality must behave consistently for dates prior to, during and after year 2000;
in all interfaces and data storage, the century in any date must be specified either explicitly or by unambiguous algorithms or inferencing rules; and the year 2000 must be recognised as a leap year.
And the Health and Safety Executive launched its Health and safety and the year 2000 problem in May of this year, which sets out a risk-based approach.
A variety of bodies have been set up to help companies cope with the Y2K problem and some deal specifically with the construction industry. The main one is the Construct IT Centre of Excellence, based at Salford University, which was set up about three and a half years ago to coordinate and promote IT research in the UK construction industry. The group is made up of several large UK contractors including Bovis, Laing, Balfour Beatty, Robert McAlpine and Kvaerner, plus management consultant KPMG.
The non-profit making organisation set up the Construction IT year 2000 task force in November last year. Last month it launched its Year 2000 Health Check, which should help construction companies evaluate the extent of the problem (see box). Construct IT's Derek Blundell says that main contractors became involved because of concerns over not only their own Y2K compliance but that of their supply chain. Their £3000 joining fee pays for the research, production and distribution of free documents like this one.
Ground Engineering has surveyed a cross-section of the geotechnical industry to gauge awareness of the problem and to assess how companies are dealing with it. Respondents were asked to comment on the level of support they felt they were getting from government and industry organisations and who they thought was legally and financially responsible for the work needed to satisfy compliance.
The results are encouraging. All of those responding say they are aware of the Y2K problem. While 33% do not think that it will affect their business, some say this is because they already have action plans in place; 6% either do not know or think it might affect them in some way.
Despite this split, 94% are doing something about it - mainly through auditing and checking compliance. Some are also asking suppliers to check their systems, usually by sending out questionnaires to these third parties. This work is costing some of the larger companies up to £250,000.
Half of the respondents feel they are getting enough support, a third say that more help is needed and 17% say that probably enough is being done. Half say that financial and legal responsibility for becoming compliant is with individual companies, with 22% saying that the computer industry should foot the bill.
While it appears that the geotechnical industry is well aware of the Y2K problem, it is clear that some companies do not think it will affect them seriously. It is still perceived as a 'PC problem' and the effect on other more everyday items that use computing power, such as telephones and faxes, has been ignored. This could lead to disruption and even loss of business to many companies. If, as predicted, a recession hits in the next couple of years and the problem is ignored, 'The New Millennium Experience' may take on an entirely different meaning.