Jim McCarthy writes about 21 Rules of Thumb that guide the development of software products inside Microsoft. There are very few surprises, but I've been involved in quite a few software products before turning in my leaf and becoming a TD, and I bet most product managers would find more of them surprising than I did. McCarthy has a very pragmatic and practical view of software engineering, and serves as a good counter-agent to the marketing hype of most software companies.
Marie Sandström has a short thesis on spoofing biometric fingerprint scanners. It involves collecting latent prints, and then casting a replica finger in gelatin. I have read about such techniques here by Tsutomo Matsumoto, and here it is two years later and the latest hardware appears no more effective.