I think this will mean that some-seq[@foo] will only work if @foo is ordered? Even accepting that restriction, some-seq[* - 2], which wants the elems (and so much produce everything), is going to be problematic with this too. I dunno if a more fruitful direction would be a Sequence overload on the postcircumfix candidates...