Today somebody posted a comment:
‘Could you write a post giving more details on why you think
checked exceptions are “the embodiment of evil”?’
So here it is.
As you can see I haven’t written a post in a long time (sorry). This post is also not going to be very comprehensive. Of course I could explain everything, but I’d rather link to another site which sums it up very nicely:
http://c2.com/cgi/wiki?TheProblemWithCheckedExceptions
Basically, I don’t want to be forced to handle exceptions. Also, having to catch checked exceptions makes code unreadable, complex and harder to change/refactor.
Of course I’m also aware that with runtime exceptions you are likely to forget to catch some exceptions, which is also very bad. That is why our application(s) catch and log all runtime exceptions and mail them to the developers. Execptions are always something very bad and should not be functional, exceptions should be… exceptions. Once they happen, all hell should break out and it should be fixed.