Broadly speaking, computer projects make three sorts of news headline. One is the ‘gee-whiz gizmo’ of fond Tomorrow’s World memory. Second is the ‘big brother’ scare story about surveillance or intrusive data-sharing.

The third category is the fiasco, the tale of a system implementation that runs late, over budget, fails to work, or a combination of all three. In one sense, it’s odd that these make news - as the vast majority of big IT-based business change projects go adrift on one or more counts, failure is a bit of a ‘dog bites man’ story. And, although most project fiasco stories concern government, there’s no evidence that public sector IT projects fail more often than private sector ones - businesses are just better at burying the bodies.

Judging by the volume and tone of readers’ comments over the past week, two big projects of importance to solicitors look like falling into the ‘fiasco’ category. They are the centralisation of civil claims processing in Salford and the mySRA online renewals process.

It is quite possible that both projects’ implementation problems will prove transitory. But the dismal history of IT fiascos shows that once the complaints have reached critical mass, these things have a habit of getting worse. The perception of disaster spreads so customers get anxious, the number of transactions soars, systems fall over, staff morale plummets, senior executives make panicky decisions, and round we go again.

We know this because it has happened time and time again, at the Passport Agency, HM Revenue & Customs' tax credits system, the Office for National Statistics and the Probation Service, to name only a few. Over the past 15 years the National Audit Office and Commons Public Accounts Committee have produced a string of reports about why big IT-based projects fail.

Thirteen years ago, a book called Crash! (by Tony Collins and David Bicknell, Simon & Schuster 1999) identified common factors in disasters. These almost invariably include overambition, unrealistic deadlines, ‘big bang’ launches, failure to engage with users of the system and custom-building rather than buying proven off the shelf technology.

Avoiding these traps is now supposed to be hard-wired into government policy. After making hay with their predecessors’ long record of disasters, the coalition government has published a new ICT strategy which is supposed to commit everyone to proven disaster-avoiding methods like starting small, using open, proven systems, and quickly correcting for mistakes. One product of this new ‘agile’ philosophy went live this week, a new government web portal built at a fraction of the cost of its predecessor and launched as a prototype so that it can be tweaked according to users’ feedback.

I hope the same principles are being applied to the crucial IT schemes now being wheeled out to support massive reforms across the courts and criminal justice system. Worryingly, there still seems to be an enthusiasm for gargantuan projects and big bang implementations to meet immovable deadlines.

Don’t say I didn’t warn you.

Follow Michael on Twitter