Invented Discoveries

• Clever code ≠ Good code.
• RDBMS > NoSql.
• Java isn't that bad.
• Shitty implementation of a good abstraction > Good implementation of a partial abstraction.
• Purity > Practicality when working with teams.
• Adding more technology is rarely good.
• Many failures start with the word "scalable".
• 95% of project managers could disappear this instant without impacting efficiency.
• Interviewing is a broken system and no one understands it.
• You probably want to build a monolith. You probably want PostgreSQL as your database.
• Stay away from test-driven development.
• To know more about the problem in the shortest time with the highest accuracy: talk to the customer.
• Premature simplification is bad security.
• Google prioritizes websites with ads.
• Invoking the function we need by passing the result as a parameter instead of invoking it where we need the result is a great way to refactor code.
• When taking opinions, center your argument around the median, not the mean.
• Different components of the system should interact via well-defined APIs with very few to zero side channels.
• Lease resources instead of allocating permanently.
• Interactions should have timeouts.
• Do not hold locks while sending notifications to avoid deadlocks.
• Never treat paths as strings.
• Repeat. Yourself. Differently.