Five Languages in a Moment

Slides for my talk “Five Languages in a Moment” in Conferencia Rails 2011 (Madrid, 14 July)

Transcript

1. Five Languages in a Moment Sergio Gil
2. Sergio Gil Pérez de la Manga @porras
4. I — Creating Software
5. I — Ruby
6. “Excusatio non petita, accussatio manifesta”
9. “Learn a new language each year”
11. 2006 Ruby
12. 2006 2007 Ruby
13. 2006 2007 Ruby -
14. 2006 2007 2008 Ruby -
15. 2006 2007 2008 Ruby - -
16. 2006 2007 2008 2009 Ruby - -
17. 2006 2007 2008 2009 Ruby - - -
18. 2006 2007 2008 2009 2010 Ruby - - -
19. 2006 2007 2008 2009 2010 Ruby - - - -
20. Ruby is awesome!!
22. THE POWER CONTINUUM
23. Highest level Lowest level
24. Highest level Lowest level Machine Code
25. Highest level Lowest level Machine Code Assembler
26. Highest level Lowest level Machine Code Assembler C
27. Highest level Lowest level Machine Code Assembler C Java
28. Highest level Lowest level Machine Code Assembler C Java Ruby
32. Abstractions are tools for the mind
37. Created by Joe Armstrong at Ericsson in 1986
38. Functional Language
39. Functional Language Concurrency Oriented
40. Functional Language Concurrency Oriented Distribution Oriented
41. Functional Language Concurrency Oriented Distribution Oriented Error Tolerant
43. Functional Programming
44. http://www.defmacro.org/ramblings/fp.html
45. Side effects free programming
46. x = x + 1
48. > X = 1.
49. > X = 1. > X. 1
50. > X = 1. > X. 1 > X = 2. ** exception error
51. Pattern Matching
53. > Point = {point, 15, 20}.
54. > Point = {point, 15, 20}. > {point, X, Y} = Point.
55. > Point = {point, 15, 20}. > {point, X, Y} = Point. > X. 15
56. > Point = {point, 15, 20}. > {point, X, Y} = Point. > X. 15 > Y. 20
57. Lists
58. Lists Head and Tail
59. Lists Head and Tail [H|T]
61. > L = [1, 2, 3, 4].
62. > L = [1, 2, 3, 4]. > [H|T] = L.
63. > L = [1, 2, 3, 4]. > [H|T] = L. > H. 1
64. > L = [1, 2, 3, 4]. > [H|T] = L. > H. 1 > T. [2, 3, 4]
65. Functions
66. double(X) -> X * 2.
67. fact(0) -> 1; fact(X) -> X * fact(X - 1).
69. sum([]) -> 0; sum([H|T]) -> H + sum(T).
70. sum([]) -> 0; sum([H|T]) -> H + sum(T). map(_, []) -> []; map(F, [H|T]) -> [F(H) | map(F, T)].
71. sum([]) -> 0; sum([H|T]) -> H + sum(T). map(_, []) -> []; map(F, [H|T]) -> [F(H) | map(F, T)]. ...
73. area({rectangle, Width, Height}) -> Width * Height; area({square, X}) -> X * X; area({circle, R}) -> 3.14159 * R.
74. area({rectangle, Width, Height}) -> Width * Height; area({square, X}) -> X * X; area({circle, R}) -> 3.14159 * R. > area({square, 3}). 9
75. class Rectangle def initialize(width, height) @width = width @height = height end def area @width * height end end class Square def initialize(x) @x = x end def area @x * @x end end class Circle def initialize(r) @r = r end def area 3.14159 * @r end end
76. def area(*args) case args[0] when :rectangle args[1] * args[2] when :square args[1] * args[1] when :circle 3.141519 * args[1] end end
77. Concurrency
78. Pid = spawn(Fun).
79. Pid ! Message.
80. receive Pattern -> Expression; Pattern -> Expression end
81. map(_, []) -> []; map(F, [H|T]) -> [F(H) | map(F, T)]. Remember map?
83. pmap(F, L) -> Parent = self(), Pids = map(fun(I) -> spawn(fun() -> Parent ! {self(), F(I)} end) end, L), gather(Pids).
84. pmap(F, L) -> Parent = self(), Pids = map(fun(I) -> spawn(fun() -> Parent ! {self(), F(I)} end) end, L), gather(Pids). gather([]) -> []; gather([Pid|T]) -> receive {Pid, Result} -> [Result|gather(T)] end.
85. xN
86. Distribution Oriented
87. More interesting things about Erlang
88. More interesting things about Erlang Error control, even distributed
89. More interesting things about Erlang Error control, even distributed OTP
90. More interesting things about Erlang Error control, even distributed OTP Web development with ChicagoBoss
91. You should try Erlang if...
92. ...you want to play with functional programming
93. ...you're interested in writing parallel programs
95. Haskell
96. Created by Simon Peyton Jones in 1990
97. Pure functional language
98. Pure functional language Lazy evaluation
99. Pure functional language Lazy evaluation Strict but powerful type system
101. mymap _ [] = [] mymap f (x:xs) = f x : map f xs
102. Haskell Type System
103. Haskell Type System Strong
104. Haskell Type System Strong Static
105. Haskell Type System Strong Static Inferred
106. Char
107. Char [Char]
108. Char [Char] String
109. Char [Char] String Integer
110. Char [Char] String Integer ([Char], Integer)
111. data BookInfo = Book Int String [String]
112. data BookInfo = Book Int String [String] myBook = Book 987987 "About Wadus" ["Fulano", "Mengano"]
114. data Color = Red | Yellow | Green | RGB Int Int Int
115. data Color = Red | Yellow | Green | RGB Int Int Int yellow = Yellow black = RGB 0 0 0
116. Type System + Pattern Matching
118. data Shape = Circle Float | Rectangle Float Float | Square Float
119. data Shape = Circle Float | Rectangle Float Float | Square Float area (Circle r) = r * 3.14 area (Rectangle w h) = w * h area (Square x) = x * x
120. data Shape = Circle Float | Rectangle Float Float | Square Float area (Circle r) = r * 3.14 area (Rectangle w h) = w * h area (Square x) = x * x circle = Circle 5 rectangle = Rectangle 2 3 square = Square 1.5
121. data Shape = Circle Float | Rectangle Float Float | Square Float area (Circle r) = r * 3.14 area (Rectangle w h) = w * h area (Square x) = x * x circle = Circle 5 rectangle = Rectangle 2 3 square = Square 1.5 -- > area circle -- 15.700001
122. data Shape = Circle Float | Rectangle Float Float | Square Float area (Circle r) = r * 3.14 area (Rectangle w h) = w * h area (Square x) = x * x circle = Circle 5 rectangle = Rectangle 2 3 square = Square 1.5 -- > area circle -- 15.700001 -- > :type area -- area :: Shape -> Float
123. More Functional Programming
124. square_all list = map (\n -> n^2) list > square_all [1, 2, 5] [1,4,25]
125. square_all list = map (\n -> n^2) list > square_all [1, 2, 5] [1,4,25] square_all list = map (^2) list
126. square_all list = map (\n -> n^2) list > square_all [1, 2, 5] [1,4,25] square_all list = map (^2) list square_all = map (^2)
127. Laziness
128. @articles = Article.where(:status => 'published')
129. @articles = Article.where(:status => 'published') <% @articles.each do |article| %> <h2><%= article.title %></h2> <p><%= article.body %></p> <% end %>
130. @articles = Article.where(:status => 'published') <% @articles.limit(5).each do |article| %> <h2><%= article.title %></h2> <p><%= article.body %></p> <% end %>
131. wadus 1 + 2
133. iforelse cond a b = if cond then a else b
134. iforelse cond a b = if cond then a else b > iforelse True (1 + 2) (3 + 4) 3
135. iforelse cond a b = if cond then a else b > iforelse True (1 + 2) (3 + 4) 3 > iforelse False (1 + 2) (3 + 4) 7
136. You should try Haskell if...
137. ...you tried the functional paradigm and want to go further
138. ...you want to see your belief “types are useless” challenged
139. ...you want to write extremelly efficient programs in a high level language
141. Common Lisp
142. Invented by John McCarthy in 1958
143. Invented by John McCarthy in 1958
145. Lots of Irritating Superfluous Parentheses
146. HOMOICONIC
148. “property of some programming languages, in which the primary representation of programs is also a data structure in a primitive type of the language itself”
149. CODE = DATA
150. (1 2 3 ("wadus" 5) (1.3 1.7))
151. (1 2 3 ("wadus" 5) (1.3 1.7)) [1, 2, 3, ["wadus", 5], [1.3, 1.7]]
152. (+ (* 3 5) (/ 10 2))
153. (+ (* 3 5) (/ 10 2)) 20
154. (1 2 3 ("wadus" 5) (1.3 1.7)) (+ (* 3 5) (/ 10 2))
155. (1 2 3 ("wadus" 5) (1.3 1.7)) (+ (* 3 5) (/ 10 2)) CODE = DATA
156. Lisp has no syntax
157. Valid Lisp code is composed of lists whose first element is the name of a function and the rest are the parameters passed
158. Macros
159. Metaprogramming
160. Metaprogramming is creating and manipulating code
161. Metaprogramming is creating and manipulating code lists
162. LISt Processing
163. (defmacro backwards (code)
164. (defmacro backwards (code) (reverse code))
165. (defmacro backwards (code) (reverse code)) (backwards ("wadus" print))
166. (defmacro backwards (code) (reverse code)) (backwards ("wadus" print)) (print "wadus")
167. You should try Common Lisp if...
168. ...you want to take a flight in a retrofuturist jetpack
169. ...you like metaprogramming
170. ...you want to try one of the most abstraction-capable and still efficient languages
171. ...you want to understand why Lisp-lovers love Lisp so much
174. Created by Rick Hickey in 2007
175. Lisp meets XXI century
176. Lisp meets the JVM
177. Lisp meets the JVM Lisp meets purelly functional programming
178. Lisp meets the JVM Lisp meets purelly functional programming Lisp meets laziness
179. Lisp meets the JVM Lisp meets purelly functional programming Lisp meets laziness Lisp meets concurrency
181. Clojure is a Lisp
182. More laziness
184. (def whole-numbers (iterate inc 1))
185. (def whole-numbers (iterate inc 1)) (first whole-numbers) ; 1
186. (def whole-numbers (iterate inc 1)) (first whole-numbers) ; 1 (take 5 whole-numbers) ; (1 2 3 4 5)
187. (def whole-numbers (iterate inc 1)) (first whole-numbers) ; 1 (take 5 whole-numbers) ; (1 2 3 4 5) (last whole-numbers) ; => CATACROKER!
189. (defn fibonacci [] (map first (iterate (fn [[a b]] [b (+ a b)]) [0 1])))
190. (defn fibonacci [] (map first (iterate (fn [[a b]] [b (+ a b)]) [0 1]))) (take 20 (fibonacci)) ; (0 1 1 2 3 5 8 13 21 34 55 89 144 233 377 610 987 1597 2584 4181)
191. STM
193. (def fulano-balance (ref 100)) (def mengano-balance (ref 80))
194. (def fulano-balance (ref 100)) (def mengano-balance (ref 80)) (dosync (ref-set fulano-balance (+ @fulano-balance 20)) (ref-set mengano-balance (- @mengano-balance 20)))
195. (def fulano-balance (ref 100)) (def mengano-balance (ref 80)) (dosync (ref-set fulano-balance (+ @fulano-balance 20)) (ref-set mengano-balance (- @mengano-balance 20))) @fulano-balance ; 120
196. (def fulano-balance (ref 100)) (def mengano-balance (ref 80)) (dosync (ref-set fulano-balance (+ @fulano-balance 20)) (ref-set mengano-balance (- @mengano-balance 20))) @fulano-balance ; 120 @mengano-balance ; 60
198. (dosync (alter fulano-balance + 20) (alter mengano-balance - 20))
199. (dosync (alter fulano-balance + 20) (alter mengano-balance - 20)) @fulano-balance ; 140
200. (dosync (alter fulano-balance + 20) (alter mengano-balance - 20)) @fulano-balance ; 140 @mengano-balance ; 40
201. More interesting things about Clojure
202. More interesting things about Clojure Web development with Compojure
203. More interesting things about Clojure Web development with Compojure
204. You should try Clojure if...
205. ...you like both Lisp and the functional approach
206. ...you need/want Java/JVM interoperability
208. Ujfalusi
211. <
212. >
213. -
215. http://github.com/porras/ujfalusi
218. You should create your own language if...
219. ...you want to learn about how languages internally work
220. ...you want to have fun doing something very geek
221. ...your favorite player leaves your team and breaks your heart
222. And the winner is...
223. ME!
224. Fun
225. New abstractions
226. New tools
227. Polyglotism
228. Go out
229. Play
230. Break something
231. Have fun
232. Thank you! :)