[Haskell-cafe] Monad of no `return` Proposal (MRP): Moving `return` out of `Monad`

[Richard A. O'Keefe]

On 8/10/2015, at 12:39 pm, Mike Meyer <mwm at mired.org> wrote:
> 
> I'm curious - do "practical" developers really feel like they have to rush out and update their tool chain whenever a new version of part of it comes out? Most of the projects I've worked on considered the language version as a fixed part of the technology stack, and almost never updated it. Even when using Python, which valued not breaking working code more than it's own zen. But changing anything that potentially affected all the code in a working project was pretty much never done, and always involved a lot of effort.

My computer spends more of its time installing new versions of Java 1.8
than it does running Java.  An application I found very useful stopped
working when Java 1.6 went away, but Java 1.6 had to go away because of
(in)security concerns.  How anyone managed to write a program in Java
1.6 that would not run in 1.8 I hope never to know.

(No, rebuilding from sources didn't work either.  Tried that.)