Modern APL in the Real World: Theory, Practice, Case Studies

on YouTube ↗01h 17m

" summary

Claim: Modern APL delivers concise, high‑performance systems and often removes the need for external libraries; demos include a client pilot that implements ML purely with APL primitives and a working deployment flow via Docker and a Jarvis REST API front end. Tooling used includes a Dyalog debugger and remote IDE for live inspection, Python/R/.NET bridges, and a component file server for large persistent arrays. Architecture favors minimal explicit layers, linear dataflow slices, and frequent refactoring, and the examples shown run successfully.

AI-written summary · about

" chapters (55)

1. 0:00Modern APL in real life LambdaConf 2020 (1 Anno Coronaviri)
2. 4:50The coding gap
3. 5:02Production/Legacy
4. 5:54Missing best practices guidelines
5. 6:50Desert Englightenment
6. 8:16APL involves Change
7. 8:28What can stay the same?
8. 8:51Recommendations
9. 9:24Tooling Architecture
10. 10:44APL Leverage Spectrum
11. 11:11Expected Scaffolding
12. 12:16High cost of removing boilerplate
13. 12:57Shift your expectations
14. 13:08Designing vs. Coding
15. 15:48Type thinking
16. 16:17Mathematical Domains
17. 16:42Write literature, not tax code
18. 17:38Affordances: C. S. Lewis vs. IRS
19. 19:30Linear vs. Non-linear
20. 19:57Emergent vs. Referent Complex Systems Analysis
21. 20:35Intentional, Incremental
22. 21:30Booleans, Number Systems
23. 22:11Increase your Working Set of Idioms
24. 23:13Read Idiom Libraries
25. 32:19Data Persistence
26. 32:55Data Analytics
27. 33:18Computational, Traditional Algorithm Libraries
28. 33:54Minimal Architecture
29. 34:34Social methods, spirit over inflexible architecture
30. 34:52Implicit architecture is the bedrock of good design
31. 38:45Slice @ domain, interface
32. 39:13Avoid internal architecture
33. 40:00Method Gap
34. 41:22Competency is your limiting factor
35. 42:04Traditional Teams
36. 42:14Specification Resolution
37. 42:34Discovery process
38. 43:17Leverage your advantage
39. 43:43What do I recommend?
40. 44:14Just start with Extreme Programming
41. 44:30APL mitigates XP's drawbacks
42. 45:37Biohacking clean code
43. 46:15Maximize strengths
44. 46:47Visual Limits
45. 47:24No unit testing
46. 48:09Minimal control flow
47. 48:22Linear, data flow
48. 49:57Version control for notes Not comments!
49. 50:58Make complexity painful
50. 51:32Force refactoring
51. 51:48Leverage Industriousness
52. 52:27Idioms, style, rules
53. 52:43Clarity, Macro Thinking
54. 53:09Don't embed fear in your code
55. 54:01Maximize simplifying affordances

" related talks

• 58mFunctional Array Funhouse Intensive - Part 2 of 5Aaron W Hsu · LambdaConf 2017
• 47mFunctional Array Funhouse Intensive - Part 4 of 5Aaron W Hsu · LambdaConf 2017
• 44mFunctional Array Funhouse Intensive - Part 5 of 5Aaron W Hsu · LambdaConf 2017
• 02h 50mFunctional Array Funhouse Intensive - Part 3 of 5Aaron W Hsu · LambdaConf 2017
• 23mPattern-Based Optimization (...)Dániel Berényi · Lambda Days 2018
• 26mLocal-firstRoland Kuhn, José Duarte · PLF 2023
• 56mPerformance in the FieldAleksandr Shushunov · HolyJS 2021 Piter
• 07mConsistent Hermetic Tooling with Actually Portable ExecutablesMatt Clarkson · BazelCon 2024