Apparently the latest version of MySQL has solved this problem:
http://www.xaprb.com/blog/2006/06/28/why-large-in-clauses-are-problematic/
But I am running PostgreSQL v8.3 and am observing generally that SELECT ... WHERE ... IN (a, b, c, ...) is much slower
thanSELECT ... INNER JOIN (SELECT a UNION ALL SELECT b UNION ALL SELECT c ...)
Why doesn't the optimizer automatically transform IN clauses to INNER JOINs in this fashion?