3 ms·
1. Unit test everything, document everything, it is ALWAYS worth the cost (at least for a minimal test + mini design/spec/javadocs and the likes) 2. Don't do p
by eranation 8y ago
1. Unit test everything, document everything, it is ALWAYS worth the cost (at least for a minimal test + mini design/spec/javadocs and the likes)
2. Don't do premature optimizations / Measure before optimizing
3. Keep your cool when asked for estimates for something with unknown requirements / unknown team. Losing your mind or temper is never worth it.
4. Beware of the silver bullet new framework, right tool for the right job, avoid they hype on one hand and learn to move on when a technology dies out. (Hi Flash/VB/Cobol/Mainframe developers, I know you still have job security for a few years... but not forever...)