It always baffled me browsers even try to download an image (or a page or whatever) I asked them to save despite fact they have already downloaded and displayed it. What I would want them to do instead is just dump it from the memory.
And this sounds particularly important in case it's about a web page which has been altered in runtime by JavaScript - I want the actual DOM dumped so I can then loaded it to display exactly what I see now.
If there was a standard checksum request within HTTP, sure. Otherwise you're going to break some workflows with this kind of aggressive caching. Maybe it should be an opt-in setting (and maybe it already is).
And this sounds particularly important in case it's about a web page which has been altered in runtime by JavaScript - I want the actual DOM dumped so I can then loaded it to display exactly what I see now.