nickbrown

I'm trying to get an ever expanding coll to predominantly access more recent entries - in other words weight it towards the higher index numbers on the fly as they come in. It feels as though it should be easy but it's got me puzzled. Any thoughts anyone?

weighting-a-coll

try a combination with the probability features of [table] maybe? Some way to have the table reflect the state of the coll, and the higher values will bang out more often than the lower ones. Lots of other probability/stats stuff there too.

Yes thanks - I think I've found an answer - and this is the principle...

but any better solutions gratefully accepted

Weighting a coll