Arne Babenhauserheide <xxxxxx@web.de> writes:

> Hi,
>
> The draft says a hash table
>
>     Has no intrinsic order for the key-value associations it contains.
>
> Does the SRFI need the guarantee that the hash table has no order?
>
> Since ordered hash tables are becoming more common (i.e. they are now
> the default in PyPy) this restriction might be too strict. Maybe you
> could loosen it up a bit to say
>
>     Can have an intrinsic order for the key-value associations it contains.
>
> Best wishes,
> Arne

The concept of "no order" covers every possible order. :-)

A programmer could not possibly tell whether the hash table really has
an order, or whether its elements just happen to be returned with the
same arbitrary order "concidentally" every time hash-table-entries or
such is used.

Such are mind twists in standardization...

Taylan