[Haskell-cafe] closed world instances, closed type families

[Henning Thielemann]

On Tue, 2 Apr 2013, Daniel Peebles wrote:

> It seems very similar to Ryan Ingram's post a few years back
> (pre-TypeNats): http://www.haskell.org/pipermail/haskell-cafe/2009-June/062690.html
> The main difference is that he introduces the "knowledge" about zero vs. suc as a constraint, and you introduce
> it as a parameter. In fact, his induction function (which is probably what I'd call it too) is almost identical
> to your switch.

The answer to his post by Miguel Mitrofanov contains a caseNat that is 
exactly my 'switch'. I see I am four years late.