Unofficial wish list status.(Jul 2008)

[Walter Bright]

Jason House wrote:
> No, I'm not.  I'm only suggesting that functions that are invariant
> and const functions are redefined to be like pure functions...
> Merging concepts together.  There's no reason for an invariant
> argument to a function require the function itself to be
> invariant/pure.

If invariance and purity were merged, then the whole system of invariant 
strings, which works great, would have to be scrapped. I don't see a 
gain that approaches that downside.