Assuming that Accepted system papers at HCI conferences typically have 3-4 person years of work behind them does it support that It is much harder to get a system paper accepted at HCI conferences than other types of contributions?