Is code quality **actually** important?

The context for the discussion was on quality vs. time-to-market

What is code quality?

• Peer reviews generally leads to good code quality
  • https://en.wikipedia.org/wiki/Broken_windows_theory
  • Linters, style checkers etc. were considered hygiene and should be done by tools, not in a review
  • Decoupling risks and architectural changes in reviews, e.g. fix risk now, ship, asses architecture as a next step or later
  • At google you need one review from someone understanding the business logic/context and one review from someone certified as "language reader" for the implementation language
• "Code quality is not getting calls on a weekend"

Various business goals setting need for code quality

• Threads was originally just a stripped down instagram, the business goal was to win market share(?) with Twitter in turmoil.
  • Code quality and other technical aspects were to be tackled later
• Solutions with actual usage care about their customer retention, customer lifetime value etc.
  • Need for extendability and maintainability, while not upsetting existing users.
• One participant worked closely with a team creating valuable PoCs in three weeks, that were then handed over for maintenance and refinement for another team if the solution gets traction
• Open source
  • Share solutions and crowdsource the evolution and maintenance

Other

• There is satisfaction in creating good quality code
• Programming languages likely to trend towards ways that enable AI code writing and reading
  • Intermediate level language?
• Prediction that front-end development role gone in 10 years
• Cobol not generatable today?
• Out of the 20 or so participants only three had clear vocabulary and process in place to distinguish between different kinds of work
  • e.g. solution core work that needs high quality vs. a bet on a new low-risk feature, just to understand would it be used at all
  • This makes code quality discussions difficult as likely all work following same standards