Get used to the idea – the (wrong) assumption you’ve used is the iconification of the window is done consistently – perhaps by binding your function to some window button, etc. But what if I use a key-binding that calls Iconify (likely – by the way). Oops, it no longer works.
There is absolutely no overhead in using FvwmEvent to achieve this. It is the best catch-all case we have. Any other suggestion against its use is inferior.
I don’t experience problems, and I use mainly the keyboard to control my wm. I am not saying you are not right (you surely are), but it just works and there is no noticeable problem.
I use win+tab to navigate all the windows, included iconified ones. And when I press Enter on the context I, the relevant window receives Iconify off, and the rest of icons re-arrange adequately. No problem as far as I can see. There is no problem in using FvwmEvent (just a few mb of ram, probably), but I see no problem in not using it either.
You’re missing the point still. There are other ways beside explicit user actions which could cause a window to iconify. If that happens, your functions aren’t called, and nothing useful happens. Binding them to FvwmEvent is still useful. I’m not going to argue the toss, but your observation of a possible use of RAM is a red-herring I’m afraid.
(Oh, and if you still don’t believe me, look at things like the StartIconic style condition as well as those applications which override the WM and will always start iconic. What do you think happens then with your functions without FvwmEvent?)