[bitc-dev] Syntax of effect types (with request for input)

[Jonathan S. Shapiro]

On Wed, 2008-08-06 at 12:10 -0400, Jonathan S. Shapiro wrote:
>   ([effect] fn ('argType0 .. 'argTypeN) 'returnType)

This syntax could equally well be:

  (fn [effect] ('argType0 .. 'argTypeN) 'returnType)

The motivation to put the effect first was that it reads better: pure
and impure are adjectives. Unfortunately it doesn't read very well for
things like MAP.

On reflection, I'm less and less convinced that 

> where "effect" may be... missing, meaning presumptively impure

is useful. It is much less backwards compatible than it initially
looked. The question, I think, will come down to a matter of
conciseness. If impure functions are the rule, then absence of effect
should mean impure.

Unfortunately we just don't have enough experience yet to know.

shap