Originally Posted by logicPwn
I didn't design it - https://crashreporterdotnet.codeplex.com/
BTW JIT isn't a crash reporter but more a debugger, it tells the current user the exception and stack trace but the user does nothing with it., and most of the time they hit continue. The crash reporter allows them to easily add a desc, even a screenshot, and send it.
To think you can forsee EVERY exception your application will have is foolish, especially while being heavily developed!
Ah right you're using that.
To be pedantic, the JIT is neither a crash reporter, nor a debugger - its a compiler
I've got projects fairly large in size and not a single exception goes unhandled on a stable build (either the exception isn't thrown full stop, or its one that can be handled). That's not because I can forsee exceptions that will go unhandled, its because I have a rigorous test suite that tests every portion of the application - unhandled exceptions are criteria for a test fail.
I just seem to think that these crash reporter things are a band-aid solution for poor development techniques. One should aim to reduce the number of exceptions to a number close to zero because exceptions hurt performance tremendously. So I guess what I am trying to say is that it is fine to use a crash reporter, but only after you've verified that you're not using it as your "test suite". Any exceptions after that will be truly exceptional and that's where the crash reporter makes sense and comes into play.
Anyway to answer your original question - I think it is much cleaner to show "1 week ago" rather than "7 days ago"