Re: Combining mutable objects with const references

On Apr 25, 1:13=A0pm, anonpost...@[EMAIL PROTECTED]
 wrote:
>
> Well, non-
> persistence certainly isn't *dangerous*. Maybe you did not mean to
> imply that the only features that need be removed are dangerous ones.

Non-persistence *is* dangerous - think subtle aliasing bugs. That is,
unless you have a linear type system that prevents them. In contrast
to your speculation, such a type system is far from trivial,
especially in a language with first-class functions. Clean seems to be
the only practical language that implements something along these
lines (if you do not count Haskell's monadic approach).

- Andreas