Fix Github ribbon and add new highlighted code example

[bstrie]

No description provided.

[bstrie]

Note the commit message in that third commit. I feel that making those variable names less cryptic is a major win for first-time readers.

[brson]

Fantastic

[nikomatsakis]

It looks great. The only caveat is that post-demoding it has to look a bit different:

fn main() {
    for ["Alice", "Bob", "Carol"].each |name| {
        let name = *name;
        do task::spawn {
            let v = rand::Rng().shuffle([1, 2, 3]);
            for v.each |num| {
                io::print(fmt!("%s says: '%d'\n", name, *num))
            }
        }
    }
}

This is because each gives an (explicit) pointer to the items in the vector now.

[nikomatsakis]

It occurs to me, as a random thought, that if we supported infallible pattern matching in the parameter list, one could write the following:

fn main() {
    for ["Alice", "Bob", "Carol"].each |&name| {
        do task::spawn {
            let v = rand::Rng().shuffle([1, 2, 3]);
            for v.each |&num| {
                io::print(fmt!("%s says: '%d'\n", name, num))
            }
        }
    }
}

which is cool, although maybe it doesn't quite look like what it means. Here the &name and &num are patterns, so they would effectively dereference name and num directly.

[bstrie]

Will the current code example be valid for 0.4?

[bstrie]

Also that second revised code example looks fine (a little hesitant to start using the sigils in the introductory code example, but not that bad), but that first revised example looks bad enough that I'd want to just rewrite the code example entirely.

[brson]

@bstrie. No, it won't be valid for 0.4.

[bstrie]

@nikomatsakis I don't suppose that infallible-matching-in-parameter-lists is so uncontroversial and easily implemented that it could arrive in time for 0.4, is it?

[nikomatsakis]

@bstrie uncontroversial, yes, easily implemented, no. It's actually not that hard...but it certainly wasn't on our priority list.

One other comment about the example: the old version relied on an implicit copy of a ~str, which i believe generates a warning (or should). It'd probably be nice to have the opening example be warning free.