Oops! OOP's Not What I Thought

(Presented at Øredev 2018, DomCode 2016)

In this talk we'll go on a historical and philosophical journey deep into the heart of object-oriented programming (OOP). Come along with me as I have my world shaken by the discovery that objects and classes are not the most important concepts in OOP: messages and late binding are. Join me as we try to peek inside the heads of Alan Kay and the other founders of the paradigm as they were developing early object-oriented languages like Smalltalk, only to discover that those "old" ideas seem strikingly relevant today. We may even feel our jaws drop as we realize that OOP and functional programming are not as different as we may have thought, and that the first object-oriented language was not created in the 60's or 70's but much, much earlier…

What awaits us at the end of this journey? At worst, we'll experience a brief crisis of faith in everything we ever thought we knew about programming. (A support group will meet after the conference.) At best, we'll shift the way we view this near-ubiquitous but oft-misunderstood programming paradigm, and walk away with new insights for how we architect and understand our code.

Transcript

1. oops!
   OOP’s not what I thought
   @AnjanaVakil
   Øredev 2018
   

   View Slide

2. @AnjanaVakil
   hi, I’m Anjana!
   - philosophy grad
   - English language teacher
   - computational linguist
   - software developer
   

   View Slide

3. @AnjanaVakil
   hi, I’m Anjana!
   live location data platform
   for developers
   mapbox.com
   

   View Slide

4. @AnjanaVakil
   hi, I’m Anjana!
   events.mozilla.org
   /techspeakers
   outreachy.org
   

   View Slide

5. @AnjanaVakil
   hi, I’m Anjana!
   The Recurse Center
   recurse.com
   

   View Slide

6. @AnjanaVakil
   hi, I’m Anjana!
   ...and I had OOP all wrong
   

   View Slide

7. @AnjanaVakil
   object-
   oriented
   programming
   objects (!)
   classes
   inheritance
   hierarchy
   C++, Java
   

   View Slide

8. @AnjanaVakil
   object-
   oriented
   programming
   objects (!)
   classes
   inheritance
   hierarchy
   C++, Java
   

   View Slide

9. @AnjanaVakil
   object-
   oriented
   programming
   ?
   

   View Slide

10. @AnjanaVakil
    Photo by Marcin Wichary from San Francisco, U.S.A. (Alan Kay and the prototype of Dynabook, pt. 5)
    [CC BY 2.0 (http://creativecommons.org/licenses/by/2.0)], via Wikimedia Commons
    Alan
    Kay
    @AnjanaVakil
    

    View Slide

11. @AnjanaVakil
    Alan
    Kay
    coiner of
    “object-
    oriented”
    @AnjanaVakil
    

    View Slide

12. @AnjanaVakil
    Alan
    Kay
    (co-)creator of
    Smalltalk
    @AnjanaVakil
    

    View Slide

13. @AnjanaVakil
    Alan
    Kay
    personal
    computing
    pioneer
    Dynabook
    @AnjanaVakil
    

    View Slide

14. @AnjanaVakil
    Alan
    Kay
    undergrad
    biology major
    @AnjanaVakil
    

    View Slide

15. @AnjanaVakil
    biological metaphors
    oops #1:
    

    View Slide

16. @AnjanaVakil
    what I
    thought
    inheritance
    taxonomy
    

    View Slide

17. @AnjanaVakil
    what it is
    cells
    organs
    

    View Slide

18. @AnjanaVakil
    procedural
    imperative
    “Clocks Gears Wallpaper” via Wallpoper (http://wallpoper.com/images/00/38/51/35/clocks-gears_00385135.jpg) @AnjanaVakil
    

    View Slide

19. @AnjanaVakil
    object-
    oriented
    “Slide of the Week: Increased WBC, April 14, 2011” via Center for Genomic Pathology
    (http://ctrgenpath.net/wp-content/uploads/2011/04/cml02.jpg) @AnjanaVakil
    

    View Slide

20. @AnjanaVakil
    

    View Slide

21. @AnjanaVakil
    

    View Slide

22. @AnjanaVakil
    big ideas
    oops #2:
    

    View Slide

23. @AnjanaVakil
    what I
    thought
    objects(!)
    classes
    

    View Slide

24. @AnjanaVakil
    Alan
    Kay
    “I'm sorry that I long
    ago coined the
    term "objects" for
    this topic because it
    gets many people to
    focus on the
    lesser idea.
    @AnjanaVakil
    

    View Slide

25. @AnjanaVakil
    Alan
    Kay
    “The big idea is
    "messaging"”
    - To Smalltalk/Squeak
    mailing list, 1998
    wiki.c2.com/?AlanKayOnMessaging
    @AnjanaVakil
    

    View Slide

26. @AnjanaVakil
    what it is
    messaging
    late-binding
    

    View Slide

27. @AnjanaVakil
    objects
    @AnjanaVakil
    

    View Slide

28. @AnjanaVakil
    messages
    “Dopamine D3 receptor” via Orientation of Proteins in Membranes Database
    (http://opm.phar.umich.edu/images/png/3pbl.png) @AnjanaVakil
    

    View Slide

29. @AnjanaVakil
    systems
    Image by The Opte Project (Internet Map)
    [CC BY 2.5 (http://creativecommons.org/licenses/by/2.5)], via Wikimedia Commons @AnjanaVakil
    

    View Slide

30. @AnjanaVakil
    thing.do(some,stuff)
    

    View Slide

31. @AnjanaVakil
    to: thing
    subject: pretty please? :D
    message: "do", some, stuff
    

    View Slide

32. @AnjanaVakil
    recipient message
    thing.do(some,stuff)
    

    View Slide

33. @AnjanaVakil
    recipient message
    thing.do(some,stuff)
    method name arguments
    

    View Slide

34. @AnjanaVakil
    Ruby
    thing.do(some,stuff)
    thing.send(:do,some,stuff)
    

    View Slide

35. @AnjanaVakil
    what does do do? how?
    thing.do(some,stuff)
    thing.do(other,stuff)
    child.do(some,stuff)
    thing.do(some,stuff)
    

    View Slide

36. @AnjanaVakil
    early
    binding
    @AnjanaVakil
    

    View Slide

37. @AnjanaVakil
    late
    binding
    @AnjanaVakil
    

    View Slide

38. @AnjanaVakil
    beings vs. behaviors
    oops #3:
    

    View Slide

39. @AnjanaVakil
    what I
    thought
    objects
    are nouns
    

    View Slide

40. @AnjanaVakil
    what it is objects
    are verbs
    

    View Slide

41. @AnjanaVakil
    data
    @AnjanaVakil
    

    View Slide

42. @AnjanaVakil
    behavior
    @AnjanaVakil
    

    View Slide

43. @AnjanaVakil
    thing.do(stuff)
    => behavior
    

    View Slide

44. @AnjanaVakil
    class Friend:
    def __init__(self, friends):
    self.friends = friends
    def is_friend_of(self, name):
    return name in self.friends
    buddy = Friend(['alan', 'alonzo'])
    buddy.is_friend_of('guy') => False
    

    View Slide

45. @AnjanaVakil
    buddy.is_friend_of('guy')
    buddy.send(:is_friend_of, 'guy')
    

    View Slide

46. @AnjanaVakil
    buddy.is_friend_of('guy')
    buddy.send(:is_friend_of, 'guy')
    buddy('is_friend_of', 'guy')
    

    View Slide

47. @AnjanaVakil
    def Friend(friend_names):
    my_friends = friend_names
    def is_my_friend(name):
    return name in friend_names
    def responder(method_name, *args):
    if method_name == 'is_friend_of':
    return is_my_friend(args)
    return responder
    buddy = Friend(['alan', 'alonzo'])
    buddy('is_friend_of','guy') => False
    

    View Slide

48. @AnjanaVakil
    def Friend(friend_names):
    my_friends = friend_names
    def is_my_friend(name):
    return name in friend_names
    def buddy(method_name, *args):
    if method_name == 'is_friend_of':
    return is_my_friend(*args)
    return responder
    buddy = Friend(['alan', 'alonzo'])
    buddy('is_friend_of','guy') => False
    

    View Slide

49. @AnjanaVakil
    def Friend(friend_names):
    my_friends = friend_names
    def is_my_friend(name):
    return name in friend_names
    def buddy(method_name, *args):
    if method_name == 'is_friend_of':
    return is_my_friend(*args)
    return responder
    buddy = Friend(['alan', 'alonzo'])
    buddy('is_friend_of','guy') => False
    

    View Slide

50. @AnjanaVakil
    def Friend(friend_names):
    my_friends = friend_names
    def is_my_friend(name):
    return name in friend_names
    def responder(method_name, *args):
    if method_name == 'is_friend_of':
    return is_my_friend(*args)
    return responder
    buddy = Friend(['alan', 'alonzo'])
    buddy('is_friend_of','guy') => False
    

    View Slide

51. @AnjanaVakil
    boundaries
    oops #4:
    

    View Slide

52. @AnjanaVakil
    what I
    thought
    imperative
    ----------
    object-otd
    ==========
    functional
    

    View Slide

53. @AnjanaVakil
    what it is
    imperative
    ==========
    object-otd
    ----------
    functional
    

    View Slide

54. @AnjanaVakil
    Smalltalk
    class True
    ifTrue: a ifFalse: b
    ^ a value
    class False
    ifTrue: a ifFalse: b
    ^ b value
    

    View Slide

55. @AnjanaVakil
    Smalltalk
    class True
    ifTrue: a ifFalse: b
    ^ a value
    class False
    ifTrue: a ifFalse: b
    ^ b value
    Lambda calculus
    TRUE := λx.λy.x
    FALSE := λx.λy.y
    

    View Slide

56. @AnjanaVakil
    Erlang
    Fred Hébert
    Learn You Some Erlang
    for Great Good
    learnyousomeerlang.com
    

    View Slide

57. @AnjanaVakil
    JS "callbags"
    André Stalz
    "Callback Heaven",
    Amsterdam JSNation 2018
    youtu.be/HssczgaY9BM
    

    View Slide

58. @AnjanaVakil
    belief structures
    oops #5:
    

    View Slide

59. @AnjanaVakil
    what I
    thought
    dogmatic
    restrictive
    

    View Slide

60. @AnjanaVakil
    what it is built for
    change
    *or should be
    

    View Slide

61. @AnjanaVakil
    simplicity
    complexity
    @AnjanaVakil
    

    View Slide

62. @AnjanaVakil
    resilience
    @AnjanaVakil
    

    View Slide

63. @AnjanaVakil
    scalability
    @AnjanaVakil
    

    View Slide

64. @AnjanaVakil
    Alan Kay:
    - The early history of Smalltalk worrydream.com/EarlyHistoryOfSmalltalk
    - Comment on Moti Ben-Ari's Objects never? Well, hardly ever!
    computinged.wordpress.com/2010/09/11/moti-asks-objects-never-well-hardly-ever/#div-comment-3766
    - E-mail to Stefan Ram on definition of OOP purl.org/stefan_ram/pub/doc_kay_oop_en
    - E-mail to Squeak list on "messaging" wiki.c2.com/?AlanKayOnMessaging
    William Cook:
    - A Proposal for Simplified, Modern Definitions of "Object" and "Object Oriented"
    wcook.blogspot.com/2012/07/proposal-for-simplified-modern.html
    - On Understanding Data Abstraction, Revisited cs.utexas.edu/%7Ewcook/Drafts/2009/essay.pdf
    Anjana Vakil, "Programming across Paradigms" youtu.be/uNPdZm5oF_E
    Andre Stalz, "Why we need callbags" staltz.com/why-we-need-callbags
    Structure & Interpretation of Computer Programs mitpress.mit.edu/sicp
    Learn You Some Erlang for Great Good learnyousomeerlang.com
    further reading & references
    

    View Slide

65. David Albert, Darius Bacon,
    Lindsey Kuper, Julia Evans, Matthew Parker, Ezekiel Smithburg,
    Juan Hernández, Andy Lee, Michael Arntzenius
    & the Recurse Center
    Team Øredev
    thank you!
    @AnjanaVakil
    

    View Slide