Re:The error model of Microsoft's new language
by asveikau on Monday December 30, 2013 @02:02PM
(#6986180)
Introducing threads and multiple processes seems slightly crazy. Remember that ideally the most common case is no failure at all and you don't want to introduce cost for that. Why not just have the point of failure unwind the stack until it sees a component boundary? Without multiple address spaces certain types of failure would blow up the whole process but that's life.
That's hand waving past any shared state but if you need that I suppose you use threads (I'm not a thread fan but I have to believe there is a thread_assert() that kills just that thread).