# Laziness is not always good

### From HaskellWiki

Generally, since Haskell is a non-strict language, you should try to make a function least strict. This is in many cases the best semantics and the most efficient implementation. However, here is an important exception from the rule:

Consider theMonoid

()

mempty = () mappend _ _ = ()

These functions are least strict, but have a subtle problem: They do not generally satisfy the monoid laws.

Remind you:mempty

mappend

forall a. mappend mempty a = a forall a. mappend a mempty = a

mappend mempty undefined = undefined

mappend mempty undefined = mempty

Is this academic nitpicking or practically relevant?

I think it is the latter one, because aMonoid

that monoid laws can be applied in every case.

A programmer expects that every occurence ofmappend mempty a

a

You might even create an optimizer rule doing this.

The above implementation ofmappend

and this gets lost when the optimization is applied.

The solution of this issue is to define

mempty = () mappend () () = () force :: () -> () force _ = ()

and write

mappend (force a) (force b)

mappend a b

If you find that example too academic, you can choose any other data type with one constructor instead.

## 1 Exercise

Find out whether it would help to definemempty = undefined

## 2 See also

- Haskell-Cafe on Laws and partial values
- Haskell-Cafe on a space leak caused by the garbage collector that did not recognize a selector-like function call
- Maintaining laziness