The scenario was that if you're in a whole heap of trouble, heading down fast (assume you're going to break something, and it's no longer possible to stay within the envelope), what is most appropriate?
Yes exactly, so let's get back to that. And what I'm saying in that regard, is that if you are in that situation, you do not want to use up the 50 % margin between design load factor and ultimate load factor to try and pull out of a dive, since that means you will exceed ultimate if you encounter a gust.
Btw, there is a lot about gust loads and gust margins, at least in FAR 23 (and therefore, I suppose, in JAR 23...). Not sure about CAR 4 ... But I'd be interested in those 3 seconds at 50% overstress? I don't understand how you could design something against such a limit, nor how you'd show compliance... And my understanding is that the airframe should be able to withstand a 50% overstress indefinitely.
Reading, in detail, and fully understanding FAR 23 has been on my to do list for a long time... Maybe it's about time to do it!
But we might be straying off topic...