One of the things the DearDiary code really is pretty poor at is dealing with the modern world. The original codebase was written in 1999, long before HTTP/1.1 was in widespread use. Which means a lot of the HTTP headers we send out aren't utilising a lot of the more modern cache control headers, whereas we rely on the older HTTP 1.0 cache control headers. Those old headers aren't as flexible, or reliable. So I've been playing with the newer ones.
If it works properly, it should speed up reading of entries you've already read recently – and it should enable DD to play nicer with ISPs proxy's too.
But it might mean I broke something – so if you notice things going weird please let me know 🙂