+1 on not specifying algorithms.  However, in the additional

procedures section a number of algorithm-specific vector

procedures are given.

+1 to Daniel's suggestion for partial sorting.  I would rather

make it a selection algorithm (top k elements unsorted),

though.  It's easy enough to sort the result with ranged

vector-sort!, and we could provide a utility for this if needed.

In the draft list-sort! is listed as both linear update and

as being required to mutate.

I think we should also consider not requiring side effects

for anything.

vector-binary-search "returns the index of the matching

element."  Presumably if no match is found either #f or -1

is returned - we should say which.

-- 

Alex