My 7 Surprises in Clojure

Slides to my talk on :clojureD 2017

Transcript

1. My 7 Surprises in Clojure Michael Pershyn 2017-02-25 @pershyn
2. My way to clojure Autoit3 2005 6 7 8 9 10 11 12 13 14 15 16 17 C# python clojure C++
3. 3 Functional programming is fun. And it makes you better programmer.
4. 4 Surprise 1. Total paradigm shift
5. 5 It is possible to happily live without • Modifying mutable variables • Using assignments • Using (jump-) control structures • Conceptualising data structures word-by-word • Non-determinism caused by concurrent threads accessing mutable state • Boilerplate code
6. 6 Space vs. Time
7. 7 Identity vs State vs Mutable State
8. 8 Surprise 2. Painless scale
9. 9 Stop using lots of things • Low-level concepts of thread, mutex and semaphores • GIL Workarounds • Debuger to debug all the related issues, caused by mutable state and non-determinism. That was real pain.
10. 10 So my expectations started to grow.
11. 11 Surprise 3. Types.
12. 12 Types, when you need them
14. 14 Thank you, repl!
15. 15 Refection warning, shall we do a type hint?
16. 16 Hint an array of Hosts
17. 17 Surprise 4. Patterns and inheritance
18. 18 Surpise 5. Remember, laziness everywhere
19. 19 Chunking the lazy sequences
20. 20 Surprise 6. History
21. 21 • Lisp 1959 • Actors 1973 • CSP (Communicating Sequential Processes) 1977 • ... History
22. 22 Surprise 7. Big Data and more
23. 23 The idea was to write 7. But good stuf didn‘t ft. • I was genuinely excited how well it works with big data • Cascalog & Storm ft perfectly with clojure • Solid base (JVM, Lisp, Libraries) • Rapid prototyping & robust production in same language • core.async • How powerful macros are • I probably stop here.
24. 24 Dear Community, Thank You! You are awesome! Keep rocking!