Conscious Decision Making

What is unconscious and what are conscious decisions. Where should we make an effort to be more conscious about our decisions when it comes to software development.

In particular:

- We need to decide more consciously whether to "Make or Buy" a piece for functionality.
- When need to build it we need to consciously decide what is going to be build, by employing the "You ain't gonna need it" (YAGNI) mindset
- When we are lacking information needed to make an informed decision, we can use "Spike and Stabilize" to learn what's needed before investing a lot of time and money

I hoped to provide some insights by telling a few stories. The stories have not been meant to be shared with the hold world, so are not included in the presentation.

Transcript

1. Conscious Decision Making Niklas Lochschmidt - @Niklas_L SoCraTes Day Berlin 2019 1
2. Climate Crisis is Real September 20th Global Climate Strike October 7th Global Rebellion #BerlinBlockieren https://fridaysforfuture.org/ https://rebellion.earth/ Get Informed - Tell the Truth - Act Now SoCraTes Day Berlin 2019 2
3. Disclaimer I am not a psychologist/economist/ researcher I could be wrong about these things They helped me, maybe they will help you as well SoCraTes Day Berlin 2019 3
4. Unconscious Decision Making Fast Intuitive Easy Susceptible to bias SoCraTes Day Berlin 2019 4
5. Conscious Decision Making Slow Logical Harder Risk of overconfidence SoCraTes Day Berlin 2019 5
6. Intuition is our Default Mode SoCraTes Day Berlin 2019 6
7. Making better intuitive decisions Learning Repetition Break systems that reinforce negative decisions > Conscious decision making SoCraTes Day Berlin 2019 7
8. Practice Self-care Stressed, depressed and hungry people make the worst decisions SoCraTes Day Berlin 2019 8
9. Topics for Today First decide whether you need to Make or Buy Build only what's needed - YAGNI Use Spike and Stabilize to learn what's needed SoCraTes Day Berlin 2019 9
10. Make or Buy SoCraTes Day Berlin 2019 10
11. SoCraTes Day Berlin 2019 11
12. Why Make It Yourself We like to build familiar things We typically underestimate Overconfidence: We can do this much better Not invented here / Clean-room design SoCraTes Day Berlin 2019 12
13. Why Buy It We don't like to do the same thing again Everyone else typically underestimates Low self-confidence: We couldn't do it better Believe in "cost-effective" outsourcing SoCraTes Day Berlin 2019 13
14. Make or Buy Wanna share a story? SoCraTes Day Berlin 2019 14
15. Questions to ask: Can it be a unique selling proposition? Does it need to delight your customers? Is it part of your core business? SoCraTes Day Berlin 2019 15
16. Prevent "Analysis Paralysis" in Buy-Decision Describe the problem you want solved Determine your criteria Decide how many options to include > Then start your research SoCraTes Day Berlin 2019 16
17. SoCraTes Day Berlin 2019 17
18. Bonus Level SoCraTes Day Berlin 2019 18
19. Contribute to Open-Source SoCraTes Day Berlin 2019 19
20. YAGNI You ain't gonna need it SoCraTes Day Berlin 2019 20
21. Start Tracking Product/Business Decisions What has been decided Why has it been decided When will we need to revise it Who has decided SoCraTes Day Berlin 2019 21
22. "YAGNI is not a justification for neglecting the health of your code base." Martin Fowler SoCraTes Day Berlin 2019 22
23. YAGNI requires more practices Constant refactoring Pair programming Tests Collective code ownership SoCraTes Day Berlin 2019 23
24. "You will need that backup eventually" What else is always needed? SoCraTes Day Berlin 2019 24
25. Bonus Level SoCraTes Day Berlin 2019 25
26. You Might Not Need <Your Favorite Tool Here> SoCraTes Day Berlin 2019 26
27. Spike and Stabilize SoCraTes Day Berlin 2019 27
28. Spike Some feature/component/code that has limited production- readiness to validate an idea Commitment by stakeholders to be reversible Shorten feedback cycle with customer Might have less tests, less documentation, different technologies Used for testing technical feasibility SoCraTes Day Berlin 2019 28
29. Delivering a Spike in Production Use user- or environment-based feature toggles Use continuous integration to prevent code divergence Don't change/disable existing tests to prevent behavior divergence SoCraTes Day Berlin 2019 29
30. Insist on a Decision Spike needs to be stabilized or removed SoCraTes Day Berlin 2019 30
31. Stabilize When Proven Worthy Add tests or rewrite Cover remaining edge cases Add documentation SoCraTes Day Berlin 2019 31
32. When did you last hack something SoCraTes Day Berlin 2019 32
33. Bonus Level SoCraTes Day Berlin 2019 33
34. Establish DevOpsoriginal in your Organization SoCraTes Day Berlin 2019 34
35. Summary First decide whether you need to Make or Buy Build only what's needed - YAGNI Use Spike and Stabilise to learn what's needed SoCraTes Day Berlin 2019 35
36. Further Reading ! The DevOps Handbook by G. Kim, J. Humble, J. Willis, and P. Debois Predictably Irrational by Dan Ariely Influence: The Psychology of Persuasion by Robert Cialdini ! Harry Potter and the Methods of Rationality by Eliezer Yudkowsky SoCraTes Day Berlin 2019 36