[nem-en] [RFC] qualified variant options

[Wiktor Zychla]

> variant A { | B | C }
>
> defines A, B and C. We consider the change, so it would define A, A.B
> and A.C.
>
> + it would unify variants with enums more
> + it would be possible to drop E_, T_ and similar prefixes, vs longer
> and more readable type names
> - it will need extra keystrokes
> - it breaks compatibility
>
> Anyway we would like to know what people think about it?

I am for it.

Frankly, I was disappointed that variant subtypes are not qualified with the 
base type name because it does not allow to have two subtypes with the same 
name. That is not ellegant.

// impossible in Nemerle 2.1, hope to see it in future
variant Tree {
    | ...
    | Empty;
}

variant List {
    | ...
    | Empty;
}

Wiktor