Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
What We Need are Standards
       in the Cloud
If a standard falls in the forest and there is nobody there to
                adopt it, is it still a standard?
My name is
Benjamin Black
And I have no standards.
Let’s talk about something interesting,
                instead:
Technology adoption.
Customer-centric view
Why adopt a technology?
Utility:
Does this solve my problem?
Interoperability:
Can I combine vendors+products
      to solve my problem?
Independence:
Can I be totally free of
   vendor lock-in?
Utility
Interoperability

    Utility
Independence

Interoperability

    Utility
These are not equal.
A hierarchy of needs.
Utility
Interoperability




Utility
Independence




Interoperability




Utility
Some problems need not go past UTILITY.
e.g., SQL (in reality)
Some problems can’t achieve UTILITY
        without INTEROP.
e.g., TCP/IP
The cycle:
1] Disruption.
MASSIVE UTILITY
(and a lot of imperfection)
Rapid adoption
  (by definition)
2] Competition.
EXPLORATION
Endless variety
Goal: Discovery
Not just of
solutions.
But of the complete problem.
Many ‘standards’
(meaning no standard)
The seeds of interop are sown.
We are currently in this stage.
3] Maturation.
FORMALIZATION
Our itch having been scratched, we turn to
          less pressing matters.
Independence
Successful standards
      formalize
what is already true.
Standards are side-effects
of successful technology.
There are no successful standards absent
      successful implementations.
All successful standards are
     de facto standards.
TANSTAAFL
Ra t e o f I n n o v a t i o n




                                 De g r e e o f S t a n d a r d i z a t i o n
Utility


Ra t e o f I n n o v a t i o n




                                 De g r e e o f S t a n d a r d i z a t i o n
Utility


Ra t e o f I n n o v a t i o n
                                                Interoperability




            ...
Utility


Ra t e o f I n n o v a t i o n
                                                Interoperability



             ...
(the innovation moves elsewhere)
Standardize too soon, and you lock to the
              wrong thing.
Unclear that it is even possible to
     standardize ‘too late’.
We have missed an important question:
What do we even mean by
      ‘standard’?
Standard
Standard

Standard specification
Standard

Standard specification

Standard committee
Standard:
What everybody uses.
Utility + Disruption



     Standard
Standard specification:
What someone hopes you’ll use.
Interoperability



Standard specification
Standard committees:
The legislative branch.
Independence



Standard committee
Concerned with:
Licensing

        Patent grants

        Open process

Interoperable implementations

        Open source

            et...
Special bonuses:
Politiks

Permanent seats

Preferential votes

Secret handshakes

       etc.
If you don’t achieve UTILITY...
Customers won’t bother.
They want their problems solved.
Standard committees don’t build.
We can only discover the right standards by
   building and exploring.
To discover the right standards, we must
   eschew standards.
Go do.
OSCON 2010 - Cloud Standards Debate
OSCON 2010 - Cloud Standards Debate
OSCON 2010 - Cloud Standards Debate
OSCON 2010 - Cloud Standards Debate
OSCON 2010 - Cloud Standards Debate
Upcoming SlideShare
Loading in …5
×

OSCON 2010 - Cloud Standards Debate

5,059 views

Published on

My slides from my debate with Sam Johnston at OSCON 2010 against the argument "What We Need are Standard".

Published in: Technology

OSCON 2010 - Cloud Standards Debate

  1. 1. What We Need are Standards in the Cloud
  2. 2. If a standard falls in the forest and there is nobody there to adopt it, is it still a standard?
  3. 3. My name is Benjamin Black
  4. 4. And I have no standards.
  5. 5. Let’s talk about something interesting, instead:
  6. 6. Technology adoption.
  7. 7. Customer-centric view
  8. 8. Why adopt a technology?
  9. 9. Utility:
  10. 10. Does this solve my problem?
  11. 11. Interoperability:
  12. 12. Can I combine vendors+products to solve my problem?
  13. 13. Independence:
  14. 14. Can I be totally free of vendor lock-in?
  15. 15. Utility
  16. 16. Interoperability Utility
  17. 17. Independence Interoperability Utility
  18. 18. These are not equal.
  19. 19. A hierarchy of needs.
  20. 20. Utility
  21. 21. Interoperability Utility
  22. 22. Independence Interoperability Utility
  23. 23. Some problems need not go past UTILITY.
  24. 24. e.g., SQL (in reality)
  25. 25. Some problems can’t achieve UTILITY without INTEROP.
  26. 26. e.g., TCP/IP
  27. 27. The cycle:
  28. 28. 1] Disruption.
  29. 29. MASSIVE UTILITY
  30. 30. (and a lot of imperfection)
  31. 31. Rapid adoption (by definition)
  32. 32. 2] Competition.
  33. 33. EXPLORATION
  34. 34. Endless variety
  35. 35. Goal: Discovery
  36. 36. Not just of solutions.
  37. 37. But of the complete problem.
  38. 38. Many ‘standards’
  39. 39. (meaning no standard)
  40. 40. The seeds of interop are sown.
  41. 41. We are currently in this stage.
  42. 42. 3] Maturation.
  43. 43. FORMALIZATION
  44. 44. Our itch having been scratched, we turn to less pressing matters.
  45. 45. Independence
  46. 46. Successful standards formalize what is already true.
  47. 47. Standards are side-effects
  48. 48. of successful technology.
  49. 49. There are no successful standards absent successful implementations.
  50. 50. All successful standards are de facto standards.
  51. 51. TANSTAAFL
  52. 52. Ra t e o f I n n o v a t i o n De g r e e o f S t a n d a r d i z a t i o n
  53. 53. Utility Ra t e o f I n n o v a t i o n De g r e e o f S t a n d a r d i z a t i o n
  54. 54. Utility Ra t e o f I n n o v a t i o n Interoperability De g r e e o f S t a n d a r d i z a t i o n
  55. 55. Utility Ra t e o f I n n o v a t i o n Interoperability Independence De g r e e o f S t a n d a r d i z a t i o n
  56. 56. (the innovation moves elsewhere)
  57. 57. Standardize too soon, and you lock to the wrong thing.
  58. 58. Unclear that it is even possible to standardize ‘too late’.
  59. 59. We have missed an important question:
  60. 60. What do we even mean by ‘standard’?
  61. 61. Standard
  62. 62. Standard Standard specification
  63. 63. Standard Standard specification Standard committee
  64. 64. Standard:
  65. 65. What everybody uses.
  66. 66. Utility + Disruption Standard
  67. 67. Standard specification:
  68. 68. What someone hopes you’ll use.
  69. 69. Interoperability Standard specification
  70. 70. Standard committees:
  71. 71. The legislative branch.
  72. 72. Independence Standard committee
  73. 73. Concerned with:
  74. 74. Licensing Patent grants Open process Interoperable implementations Open source etc.
  75. 75. Special bonuses:
  76. 76. Politiks Permanent seats Preferential votes Secret handshakes etc.
  77. 77. If you don’t achieve UTILITY...
  78. 78. Customers won’t bother.
  79. 79. They want their problems solved.
  80. 80. Standard committees don’t build.
  81. 81. We can only discover the right standards by building and exploring.
  82. 82. To discover the right standards, we must eschew standards.
  83. 83. Go do.

×