Pangeo Developers Meeting: Pangeo Principles

Transcript

1. The story of the
   Informatics Lab
   Jacob Tomlinson
   Technologist
   Meg Fitzsimons
   Designer
   JACOB TOMLINSON
   

   View Slide

2. View Slide

3. View Slide

4. View Slide

5. View Slide

6. PRINCIPLES
   

   View Slide

7. WHY HAVE PRINCIPLES?
   ● We are facing a complex problem
   ● We are building a complex solution
   ● Principles are tools to help you make judgement calls
   ● Principles guard against brittle and inflexible design
   ● Principles help you avoid being seduced by short terms wins
   ● Principles don’t commit to specific tools
   

   View Slide

8. “all models are wrong, but some are useful”
   George Box
   

   View Slide

9. Functionality
   What should Pangeo do?
   Implementation
   How could we do those things?
   CATEGORIES OF PRINCIPLES
   Engineering
   How can we build them well?
   

   View Slide

10. FUNCTIONALITY
    

    View Slide

11. Interactivity
    

    View Slide

12. Data
    discovery
    

    View Slide

13. Publication
    And
    Communication
    

    View Slide

14. Reproducibility
    

    View Slide

15. Expressiveness
    

    View Slide

16. Franchisable
    

    View Slide

17. Empower
    people to go
    off piste
    

    View Slide

18. Implementation
    

    View Slide

19. Lazy Data
    Access
    

    View Slide

20. Just in time
    computation
    

    View Slide

21. Scalable
    Volatile
    Compute
    

    View Slide

22. Parallel
    data access
    

    View Slide

23. Thin client
    

    View Slide

24. Engineering
    

    View Slide

25. Modular
    

    View Slide

26. Open Source
    

    View Slide

27. Specialise
    late
    

    View Slide

28. Undifferentiated
    heavy lifting
    

    View Slide

29. Writing code is a
    last resort
    

    View Slide

30. Functionality
    Interactivity
    Reproducibility
    Data Discovery
    Publication and communication
    Expressiveness
    Franchisable
    Empower people to go off piste
    Implementation
    Lazy data access
    Just in time computation
    Scalable Volatile Compute
    Parallel Data Access
    Thin client
    PANGEO PRINCIPLES
    Engineering
    Modularity
    Open Source
    Specialise Late
    Undifferentiated Heavy Lifting
    Writing code is a last resort
    

    View Slide

31. OUR NEXT CHALLENGES
    

    View Slide

32. Geo scientists
    shouldn’t have to be
    system engineers
    

    View Slide

33. View Slide

34. View Slide

35. www.informaticslab.co.uk
    @_jacobtomlinson
    THANK YOU - QUESTIONS?
    

    View Slide