I thought this was a nicely constructed puzzle with a good deal of thematic material. Tougher than previous puzzles this year, but not terribly hard.I was surprised at the double unches. S_pugh refers to one example (which could have been avoided by changing one cell) but I think there are three more, unless I've overlooked a crossing answer. Three of them have very straightforward clues, so perhaps the editors raised no objection on those grounds. I thought the misprints, though not hard to spot, were often rather inventive.
I found confirming some of the titles, and one name, a bit frustrating at the end. Three titles are obscure and difficult to trace on the internet even after consulting comprehensive lists. Still, the lovely final PDM cleared up several uncertainties and exposed a wrong entry on my part.
Re comments on 4328, I cannot see anything ambiguous in the preamble, which made it perfectly clear that there was an additional instruction. I was also one who deleted the wrong rows and columns initially, but confronted by nonsense I tried the alternative, which revealed what I expected to see.