I got a call from Rolls-Royce today, which woke me up. The last one was about five years ago. It was Peter Morgan, the engine manufacturer's director of corporate affairs.
Perhaps, I hoped, he was finally going to give me some information relevant to the ongoing inquiry into the Qantas QF32 accident - the one in which the No 2 Trent 900 engine on one of its A380s suffered uncontained failure, severely damaging the airframe and aircraft systems.
No such luck. Mr Morgan wanted to persuade me to withdraw information published in Flight International and on Flightglobal, on the grounds that it was misleading.
We had reported that the fault which caused the engine failure only affected the earlier Trent 900 versions, the A and B variants, but not the C. Morgan said that implied that Rolls knew about the weakness, and had eliminated it in the C version. But, he told me: "It is not true that we knew about a problem in the A and B versions of the engine and went on to correct it in the C version. There has been no design change relevant to this failure between A, B and C versions of the engine."
So, I asked him, is the C vulnerable to the same failure as the A and B? No, he said, the C is not vulnerable to that failure.
I asked what the difference was between the two. What had Rolls introduced which rendered the C resistant to a QF32-type failure? Morgan consulted his colleagues at Rolls and decided not to comment.
I had a pretty frank - but scrupulously polite - exchange of views with Morgan.
Why will Rolls never answer reasonable questions? They have always been the same (and for me, always is many years). And I am not alone among my colleagues in finding Rolls a closed book.
Why will Rolls not provide information about what they are doing to rectify the QF32 problem in the Trent 900 series? They think it is enough to inform us that they are doing something, but they will not discuss what it is. They discuss it with their "clients", the airlines and the airframe manufacturer, but not elsewhere.
So it's a case of "Not in front of the children", apparently.
I put it to Morgan that Rolls' ultimate clients are the fare paying passengers who are, right now, trying to make an intelligent decision about whether to ask Qantas, or Singapore Airlines, or Lufthansa to put them on one of the Boeing 747 departures instead of an A380.
The gibe did not work.
But the passengers have nothing to worry about, he told me, adding that proof of the market's confidence is that Rolls-Royce shares have climbed back to the price they had before QF32.
So that's alright, then.
PS: Morgan has just rung me to say that, if he said that the C version was not vulnerable to the QF32 fault like the A and B versions are, he should not have done.
The Rolls-Royce message remains clear: providing information is dangerous, so don't do it.