After an incident in which the Department of Social Security overpaid 112,000 benefits claimants by £10.5m because of a series of IT-related problems, neither the supplier nor the customer were held to blame.
An internal inquiry following the incident did not mark the department, the main services supplier Electronic Data Systems system, or any other third parties as being responsible.
The inquiry has been conducted in secret. The results are being kept secret. The costs of resolving the problems are also being kept secret, including the expense of collecting the income support overpayments from 112,000 claimants.
It is also likely that some of the £3m still outstanding from the overpayments may never be recovered.
A few months earlier the DSS scrapped a debt accounting and management system. The failure may never have come to the attention of parliament if not for the persistent questioning of former labour welfare minister Frank Field.
In response to his questions, the DSS told Field, "Both parties agreed to cease further development work because it had become too complex, but we did feel that we derived value from the exercise, to the tune of the investment we had made."
But it has refused to say how much it paid the supplier Electronic Data Systems, or how much of its management time it invested in the project, or how and why it derived value from the exercise. It told Field that the costs of the project are "commercial in confidence".
The DSS is no stranger to IT failures and problem projects. In the 1980s, a system called Camelot that was designed to computerise welfare benefits was scrapped with losses to the taxpayer of about £6m.
The project's replacement was called the Operational Strategy. Parliament was told it would cost £713m. This was a figure, the department told the House of Commons Public Accounts Committee, that was "unlikely to rise in real terms".
In fact, the project ended up costing £2.6bn.
In the 1990s, the DSS stopped funding, and then abandoned, a £25m Analytical Services Statistical Information System project. The department refused to say how much of its legal and management costs were not recovered from the supplier Cap Gemini, or what, if any, lessons it had learned from the exercise.
With little or no accountability for its failures, can the DSS' senior executives be expected to learn the lessons from earlier projects?
In the private sector major IT disasters are usually kept secret, and sometimes the losses run into hundreds of millions of pounds, as in the failure of the Stock Exchange's Taurus project.
But there is usually accountability for the losses in organisations that have shareholders. After the Taurus project collapsed, Peter Rawlins, the chief executive of the Stock Exchange resigned and the Bank of England, in designing the replacement for Taurus, was careful to learn from the mistakes of the past.
It could also be argued that major failures rarely happen successively in the private sector. For example, Save & Prosper's then chief executive Peter Roney said after experiencing a major IT disaster that his company would in future embark on much shorter-term projects that would be expected to deliver business benefits within six months and certainly no longer than a year.
Such edicts are rare or unknown in the public sector. Every project tends to be treated as unique.
On the other hand, the DSS' early scrapping of the debt accounting and management system marks a positive departure from some earlier disasters.
In the past, overly ambitious projects have continued, and eventually lost huge sums, because earlier cancellation was deemed politically inexpedient.
But abandoning a project within a culture of secrecy usually leaves many questions unanswered.
Was the specification a true reflection of the size and complexity of the task? Were the likely costs set unrealistically low and the business benefits unrealistically large to win approval for the project? How did the public sector end up paying their suppliers despite the scrapping of systems that the suppliers said initially were achievable?
And to what extent did the suppliers try to dissuade the customer from pressing ahead with an overly ambitious project? Or did the suppliers give encouragement? When a project is abandoned are the costs of failure born equally by supplier and customer, or mainly by the customer?
Yet, if an organisation's culture encourages only positive comments and discourages criticism, these questions will rarely, if ever, be raised, let alone answered.
It may also be argued that this culture of secrecy, defensiveness and lack of accountability is an immutable gene that marks the way that UK central government departments, NHS organisations, and some local councils have evolved.
If so, the stage is set for joint business and IT project disasters to continue, and the lessons from the past to be forgotten. Or, more likely, the lessons will simply be deemed irrelevant to the particular project in hand.