The added complexity is because my procedure always adjusts the sequence number so that it is continuous and zero based. Your proc works fine but you won't end up with continuous sequence numbers starting at zero. If you had say thirty items in the list and you randomly deleted fifteen, you would still have gaps between your sequence numbers (as you may have deleted all the rows sequenced between 5 and 15). Your proc would indeed get the sequencing correct but still with the gaps in the numbers, whereas mine would remove the gaps. Basically, mine takes the actual row count into effect, so you get back a sequence number range equal to zero to the row count -1. Admitedly this may be a bit over the top but I hate gaps in sequence numbers and know these will occur due to deletes. So I don't just act on the existing sequence numbers, I do the required move and then adjust them to be continuous.

Good looking proc though. Nice to see classy SQL.

Cheers, James