Working people to death to ship any one feature or product is a poor strategy, as it reduces the capacity to ship the next feature or product (burn out, build-up of bad rush practices). It’s far more important to have a system for shipping that improves over the long term than one that heroically manages one monster push.

Second is Jeff Bezos’ idea: “What’s not going to change over the next 10 years?” If you’re going to stick around for decades, you’re better off making investments in things that’ll pay off for a very long time. It applies both to software and peopleware.