Gavin King

> So in that context, a `[ 1, 1 ]` would not be considered to be a `Set`..? Well, I mean, *mathematically*, a set is a value with just one...

> override in `Range` with versions checking whether the "other" collection is also a `Range`, and fall back to the `Collection` implementation if not? They could be overridden on `Sequential`...

I've implemented this idea (it was really straightforward) and it seems to be working well. I'll push it to a branch later.

I've pushed my work to the 7435 branch. Now we need to see if there is really a consensus in favor of this. Feedback, please!

> [ 1, 1 ] is indeed **not** considered to be a mathematical set, because a set is a collection of **distinct objects**. This isn't really correct, at least not...

> if a `Set` doesn't specify anymore if it contains distinct objects or not, what's the difference between a `Set` and a `Collection` then? There is no suggestion to change...

@Voiteh would it work for you if these were *declaration* references, i.e. ```ceylon switch (dec) case (value PersonDto.name) { .... } case (value PersonDto.age) { .... } ```

@Voiteh well generally we don't consider a `switch` to be exhaustive unless you've *explicitly* declared that the list of members is exhaustive using an `of` clause. So I don't think...

@Voiteh well, here's an example of a class with an enumerated list of elements: ```ceylon class Suit of hearts | diamonds | clubs | spades { String name; shared new...

So the difference in the compiled code is: ```js var obj1$ = this; return (true ? obj1$.Item$obj1$c_make : obj1$.Item$obj1$c_make)(); //bad ``` Vs: ```js var obj1$ = this; return (true ?...