Nick now thinks that he “got it wrong”. Perhaps he would also like to acknowledge the possibility that the governors of Winchester may have made a similar error when they appointed him.

Actually, the world of education needs reminding that it greeted the ASA level changes with almost universal approval - twice.

In 1996, when Ron Dearing published his report on 16-19 qualifications, it was recognised as promoting breadth in sixth-form studies and addressing the high drop-out from A-level courses while maintaining standards. The report gained a favourable reception. I know: I was partly responsible for editing and launching it.

In 1997, the new government was cautious so it carried out its own consultation, getting more or less the same answers to more or less the same questions. In mid-1999, it approved the changes - but with the original implementation date of 2001.

So what went wrong? First, in the hurry to implement the changes no-one stepped back to ask what would happen in schools (Ron would never have made that mistake). Second, no-one thought through the relationship between modular A-levels and the new AS, so creating significant extra burdens through duplication. This happens time after time in education - the new is laid on top of the old, not in place of it.

Overall, the lesson is that you can’t rush implementation.

QCA has done a reasonable job of damage limitation. But what is really needed is a radical review of the purpose and extent of exams in secondary education with a clear focus on fewer tests but driving higher achievement.

Until we get a review which looks at radical options and tries to give real-world answers to the purposes of all this testing, we are likely to be left with piecemeal reforms rather than a truly integrated and simplified vision.

Tony Millns (former assistant chief executive QCA) 32 Cyprus Road, Cambridge