Away day
0 likes604 views
The document discusses a chaotic software development process involving excessive code, a high number of tests, and numerous dependencies, highlighting the complexities of managing a large codebase. It reflects on the challenges faced by development teams, including issues of trust within the offshore team and the potential project cancellation. The author emphasizes the need for practical refactoring and prioritizing working software over perfection, suggesting a shift in approach to focus on delivery and iterative improvement.
Away day
- 2. why?
- 3. l iver
- 4. we
- 8. the mission
- 9. find . -type f ( -name '*.css' -o -name '*.js' -o - name '*.java' -o -name '*.jsp' -o -name '*.html' ) -print0 | xargs -0 cat | wc -l
- 10. find . -type f ( -name '*.css' -o -name '*.js' -o - name '*.java' -o -name '*.jsp' -o -name '*.html' ) -print0 | xargs -0 cat | wc -l 558982 lines of code
- 11. grep -R ‘<bean’ . | wc -l
- 12. grep -R ‘<bean’ . | wc -l 684 spring beans
- 13. grep -R ‘class’ . | wc -l
- 14. grep -R ‘class’ . | wc -l 3096 java classes
- 15. grep -R ‘@Test’ . | wc -l
- 16. grep -R ‘@Test’ . | wc -l 1774 junit4 tests
- 17. grep -R ‘public void test’ . | wc -l
- 18. grep -R ‘public void test’ . | wc -l 1603 junit3 tests
- 19. our nice tech stack
- 20. integration 40 queues 157 fields 20 schemas 20 cores ~ 30 offline jobs 5 webservices
- 22. ~le tests
- 23. ~le tests • 2777 junit tests
- 24. ~le tests • 2777 junit tests • 360 functional tests
- 25. ~le tests • 2777 junit tests • 360 functional tests • ruby 1.8.6 (no bundler!)
- 26. ~le tests • 2777 junit tests • 360 functional tests • ruby 1.8.6 (no bundler!) • windows xp
- 27. ~le tests • 2777 junit tests • 360 functional tests • ruby 1.8.6 (no bundler!) • windows xp • monkeypatches for IE/windows
- 28. ~le tests • 2777 junit tests • 360 functional tests • ruby 1.8.6 (no bundler!) • windows xp • monkeypatches for IE/windows • 4 hours build
- 29. teams
- 30. 4 teams 35 devs
- 31. 6 engineers
- 32. 6 managers
- 33. pigeons!
- 34. many of them
- 35. first story product feed import
- 36. IHateYouForeverImpl • 2272 lines • 32 dependencies • getItemStatus (3 setters)
- 37. IHateYouServiceImplTest 78 lines of test setup
- 38. loadAllProducts.xml 26 pages
- 39. ProductDTO.java • 86 attributes • + getters / + setters • equals() => 400 lines
- 40. let’s throw it all away and start from scratch!
- 43. ok! let’s AMAZE this codebase
- 44. pague o aluguel!
- 45. massive cleanup • intensive refactoring • pair programming • rant! whining! crying!
- 46. 6 weeks later...
- 47. zoom++
- 48. fuuuu • silly refactoring (speculative) • bugs in prod
- 50. todos ‘chora’
- 51. todos ‘chora’ • little trust on the offshore team
- 52. todos ‘chora’ • little trust on the offshore team • the project almost gets cancelled
- 53. todos ‘chora’ • little trust on the offshore team • the project almost gets cancelled • IM leaves
- 54. todos ‘chora’ • little trust on the offshore team • the project almost gets cancelled • IM leaves • tech lead leaves
- 55. we were alone :’(
- 56. new approach
- 60. purism must stop
- 61. right software > software right
- 62. is it worth refactoring? • does it block another story? • does it block delivery?
- 63. speculation “this is going to impact on story #125 of the next iteration” facts x speculation software is inherently fuzzy
- 64. legacy code is working code http://favim.com/image/467371/
- 65. sweep it under the carpet façades parallel class hierarchy
- 66. tolerate heresy
- 68. reduce the feedback cycle is there anyone using it?
- 69. tools
- 73. what do they wanna hear?
- 74. is it possible not to test?
- 77. in the end...
- 79. the client
- 80. is this success a consequence of the initial effort?
- 81. sim mas não
- 82. here’s my number nettofarah@gmail.com @nettofarah vgomes@thoughtworks.com @vvgomes so call me, maybe
- 83. thank you!
Editor's Notes
- #2: \n
- #3: \n
- #4: \n
- #5: \n
- #6: \n
- #7: \n
- #8: \n
- #9: \n
- #10: \n
- #11: \n
- #12: \n
- #13: \n
- #14: \n
- #15: \n
- #16: \n
- #17: \n
- #18: \n
- #19: \n
- #20: \n
- #21: \n
- #22: \n
- #23: \n
- #24: \n
- #25: \n
- #26: \n
- #27: \n
- #28: \n
- #29: \n
- #30: \n
- #31: \n
- #32: - find . -name '*.java' | grep -v 'generated' | xargs wc -l | sort\n
- #33: - find . -name '*Test.java' | grep -v 'generated' | xargs wc -l | sort\n
- #34: \n
- #35: \n
- #36: \n
- #37: - o sistema &#xE9; gigante.\n- ultra complexo\n- reescrever significa fazer engenharia reversa\n- caso do netscape\n
- #38: \n
- #39: \n
- #40: - seguir a ideia do seu barriga\n - cobrar a d&#xED;vida (no matter what)\n - atitude extrema\n
- #41: - refactoring como se n&#xE3;o houvesse amanh&#xE3;\n- pareamento o tempo todo, v&#xE1;rias ideias e brainstorms sobre como melhorar o design\n- reclama&#xE7;&#xE3;o sobre o c&#xF3;digo o tempo todo\n
- #42: \n
- #43: - ainda estou dando uma pr&#xE9;via ali do futuro\n
- #44: \n
- #45: \n
- #46: \n
- #47: \n
- #48: \n
- #49: \n
- #50: \n
- #51: \n
- #52: \n
- #53: \n
- #54: \n
- #55: \n
- #56: \n
- #57: \n
- #58: - refatorar gerava satisfa&#xE7;&#xE3;o para o time, mas n&#xE3;o para o neg&#xF3;cio\n - esfor&#xE7;o x dano\n
- #59: \n
- #60: - o stalone &#xE9; o &#xFA;nico cara que com 60 anos derruba um helic&#xF3;ptero usando uma moto\n- assistam e vejam a piada do chuck norris\n- esque&#xE7;am os caras da esquerda.. os legados s&#xE3;o os velhos aqui do lado direito\n
- #61: - n&#xE3;o &#xE9; pq todo o c&#xF3;digo relacionado a produtos est&#xE1; numa classe que voc&#xEA; precisa continuar aumento o que acontece por l&#xE1;.\n- tente criar uma classe paralela e resolver o problema por l&#xE1;.\n- empacota tudo numa fa&#xE7;ade e toca o pau\n
- #62: \n
- #63: \n
- #64: \n
- #65: \n
- #66: \n
- #67: \n
- #68: \n
- #69: \n
- #70: \n
- #71: \n
- #72: \n
- #73: \n
- #74: \n
- #75: \n
- #76: \n
- #77: \n
- #78: \n
- #79: \n
- #80: \n