On 11 Oct 2025, at 23:32, Daphne Preston-Kendal <xxxxxx@nonceword.org> wrote:

>> Also,
>> as only one container can be involved in unordered contexts, allow an
>> arbitrary number of seeds (so that the procedure called iteratively
>> can yield more than one value).
>
> I don’t understand this suggestion. An unordered fold isn’t guaranteed to be randomly ordered; it only might be.

Oh, duh, you meant the seed which is the initial value for the accumulator, not the random seed of the hash function.

I agree this would be useful but it’s too much of an innovation to stick in at this late stage in the SRFI process. It could be an extension later.

Daphne