[SciPy-user] Future directions for SciPy in light of meeting at Berkeley