hello *, i read on the wiki about using recip(rord(...)...) to boost recent documents with a date field, does anyone have a good function for doing something similar with unix timestamps?
if not, is there a lot of overhead related to counting the number of distinct values for rord() ? thx much --joe
