>
>I had a really big problem with some queries that were ran on MySQL (perhaps two years ago) and that IN clause just killed the process. It was a huge problem for the client and lots of frustration until the cause was found. Since then I prefer Exists to be on the good side. Is there a situation where IN() with subquery would absolutely give you a better result compared with Exists()? Then I would rethink my strategy.
Question - are you saying your experiences with bad performance using IN were with the product MySQL?