2017-01-31 7:45 GMT+01:00 Saurabh Nanda <saurabhnanda@gmail.com>:
We want this at my company too, so we don't write fall-through cases.

That's a short-term workaround, yes. However it doesn't fit into Haskell's overall story of language-enforced correctness.

I don't see this as a workaround, this is *the* way to go IMHO. Using '_' is saying: "I know what I'm doing here, for all eternity, trust me...", so you get what you ask for. Unless you don't care about an argument at all, using '_' is counterproductive for maintenance. But that's the usual tension between being able to write something down quickly which works *now* and writing something which will be maintained for a long time by lots of people. So '_' itself is fine, but you should be aware of what kind of SW you are writing. Implicitness will always hurt you sooner or later, it is only a matter of time, and '_' has a very implicit flavor.

For exactly this reason, using C++'s "default" case was banned in the last 2 companies I've worked for, and this turned out to be very beneficial: Finding all the places where a "default" was not really the default anymore hit us several times and resulted in actual bugs in released SW. Banning '_', just like Christopher mentioned, seems to be a sensible approach.