Reproducible science: the good, the bad, the ugly and the untold

Slide 1

Slide 1 text

Tania Allard, PhD [she/her] @ixek Developer Advocate at Microsoft Reproducible science: the good, the bad, the ugly and the untold PyCon UK 2019 bit.ly/PyConUK-reproducibility

Slide 2

Slide 2 text

2 A bit about me Alan Turing Institute Industrial Fellow I am a recovering researcher Ex RSE – UK RSE Society Trustee JOSS editor @ixek

Slide 3

Slide 3 text

3 I spend a lot of time thinking about reproducible and open science @ixek

Slide 4

Slide 4 text

4 I spend a lot of time helping researchers to make research reproducible and open @ixek

Slide 5

Slide 5 text

5 I spend a lot of time helping data scientist to make machine learning reproducible and transparent @ixek

Slide 6

Slide 6 text

https://doi.org/10.5281/zenodo.3332808 It’s a spectrum

Slide 7

Slide 7 text

7 https://doi.org/10.5281/zenodo.3332808 @ixek

Slide 8

Slide 8 text

8 https://doi.org/10.5281/zenodo.3332808 @ixek

Slide 9

Slide 9 text

9 https://doi.org/10.5281/zenodo.3332808 @ixek

Slide 10

Slide 10 text

10 https://doi.org/10.5281/zenodo.3332808 @ixek

Slide 11

Slide 11 text

11 https://doi.org/10.5281/zenodo.3332808 @ixek

Slide 12

Slide 12 text

12 Reproducible != Open @ixek

Slide 13

Slide 13 text

THE GOOD

Slide 14

Slide 14 text

No content

Slide 15

Slide 15 text

No content

Slide 16

Slide 16 text

16 Software has changed the world @ixek

Slide 17

Slide 17 text

17 Software is changing the world… and research. @ixek

Slide 18

Slide 18 text

18 Software Sustainability Institute Survey – do researchers use software? @ixek https://slides.com/simonhettrick/why-recognising-scientific-software-experts-is-key-to-open-science#/2/1

Slide 19

Slide 19 text

19 Software Alliance @ixek https://slides.com/simonhettrick/why-recognising-scientific-software-experts-is-key-to-open-science#/2/1

Slide 20

Slide 20 text

No content

Slide 21

Slide 21 text

21 A stellar example NASA, https://flic.kr/p/tJbJf5.

Slide 22

Slide 22 text

https://iopscience.iop.org/journal/2041-8205/page/Focus_on_EHT

Slide 23

Slide 23 text

https://twitter.com/sweichwald/status/1116430285342695424

Slide 24

Slide 24 text

24 @ixek q First M87 Event Horizon Telescope Results. III. Data Processing and Calibration q A series of 6 papers published in April 2019 q Incredible long term international collaboration (200+ scientists, 60 institutes, 18 countries, 6 continents) https://doi: 10.3847/2041-8213/ab0c57

Slide 25

Slide 25 text

Open source is EXTREMELY important for research

Slide 26

Slide 26 text

26 @ixek

Slide 27

Slide 27 text

27 @ixek Open tools and open infrastructure

Slide 28

Slide 28 text

No content

Slide 29

Slide 29 text

THE BAD

Slide 30

Slide 30 text

30 What we say we want from research: - Innovation - Openness - Collaborations @ixek

Slide 31

Slide 31 text

31 @ixek

Slide 32

Slide 32 text

32 @ixek

Slide 33

Slide 33 text

33 What academia actually demands @ixek

Slide 34

Slide 34 text

The research cycle - simplified @ixek

Slide 35

Slide 35 text

The research cycle - simplified @ixek

Slide 36

Slide 36 text

36 How can research be transparent and reproducible if some parts are hidden? @ixek

Slide 37

Slide 37 text

37 An article about computational science in a scientific publication is not the scholarship itself, it is merely advertising of the scholarship. The actual scholarship is the complete software development environment and the complete set of instructions which generated the figures Buckheit and Donoho (paraphrasing John Claerbout) Wavelab and Reproducible Research 1995 @ixek

Slide 38

Slide 38 text

https://www.nature.com/news/1-500-scientists- lift-the-lid-on-reproducibility-1.19970 https://www.slideshare.net/JimGrange/the- reproducibility-crisis-in-psychological-science-one- year-later https://xkcd.com/882/

Slide 39

Slide 39 text

39 There is a reproducibility crisis! @ixek

Slide 40

Slide 40 text

40 There is a reproducibility chronic problem! @ixek

Slide 41

Slide 41 text

THE UGLY

Slide 42

Slide 42 text

@ixek

Slide 43

Slide 43 text

Researchers won’t change practises even if invited to. 43

Slide 44

Slide 44 text

Not without the right incentives

Slide 45

Slide 45 text

45 Let’s look at some examples @ixek

Slide 46

Slide 46 text

46 https://statmodeling.stat.columbia.edu/2013/04/16/memo-to-reinhart- and-rogoff-i-think-its-best-to-admit-your-errors-and-go-on-from-there @ixek

Slide 47

Slide 47 text

47 https://statmodeling.stat.columbia.edu/2013/04/16/mem o-to-reinhart-and-rogoff-i-think-its-best-to-admit-your- errors-and-go-on-from-there @ixek https://www.bbc.co.uk/news/magazine-22223190

Slide 48

Slide 48 text

https://retractionwatch.com/2019/01/04/japanese-stem-cell-fraud- leads-to-a-new-retraction/ https://www.theguardian.com/science/2015/feb/18/haruko- obokata-stap-cells-controversy-scientists-lie

Slide 49

Slide 49 text

https://retractionwatch.com/2019/08/13/doing-the- right-thing-psychology-researchers-retract-paper- three-days-after-learning-of-coding-error/

Slide 50

Slide 50 text

THE UNTOLD

Slide 51

Slide 51 text

Reproducibility is good right? Why would we be talking about it if not?

Slide 52

Slide 52 text

52 Why reproducibility matters @ixek Protects against bad actors Helps with correctness Helps ensure robustness Makes it easier to collaborate

Slide 53

Slide 53 text

53 Why reproducibility matters @ixek Leads to progress Enables strong baselines Is necessary for extensibility Makes you trustworthy

Slide 54

Slide 54 text

https://doi.org/10.5281/zenodo.3332808

Slide 55

Slide 55 text

55 That means supporting open infrastructure and open software @ixek

Slide 56

Slide 56 text

56 To help with the last mile problem @ixek

Slide 57

Slide 57 text

57 https://doi.org/10.5281/zenodo.2747640

Slide 58

Slide 58 text

58 https://doi.org/10.5281/zenodo.2747640

Slide 59

Slide 59 text

59 https://doi.org/10.5281/zenodo.2747640

Slide 60

Slide 60 text

60 https://doi.org/10.5281/zenodo.2747640

Slide 61

Slide 61 text

61 https://doi.org/10.5281/zenodo.2747640

Slide 62

Slide 62 text

62 https://doi.org/10.5281/zenodo.2747640

Slide 63

Slide 63 text

63 Are we making the best use of our public infrastructure? @ixek 63

Slide 64

Slide 64 text

64 But the last mile is the hardest one! @ixek

Slide 65

Slide 65 text

65 People are the hardest part of reproducible science @ixek

Slide 66

Slide 66 text

COMPETITIVE

Slide 67

Slide 67 text

https://doi.org/10.5281/zenodo.3332808

Slide 68

Slide 68 text

68 Researchers are not developers… @ixek

Slide 69

Slide 69 text

https://doi.org/10.5281/zenodo.3332808

Slide 70

Slide 70 text

70 @ixek Software engineering Research

Slide 71

Slide 71 text

71 @ixek Researcher Software engineer Software engineering Research

Slide 72

Slide 72 text

72 @ixek Researcher Software engineer Researcher developer Research Software Engineer Software engineering Research

Slide 73

Slide 73 text

73 Researchers are not data managers… @ixek

Slide 74

Slide 74 text

https://doi.org/10.5281/zenodo.3332808

Slide 75

Slide 75 text

75 People are the hardest part of reproducible science reproducibility and software citation @ixek

Slide 76

Slide 76 text

76 @ixek Ø Citations are academic currency (whether they should be or not!) Ø They’re the best way we have to endorse good work. Ø We should be citing the software we use.

Slide 77

Slide 77 text

No content

Slide 78

Slide 78 text

Fighting the good fight

Slide 79

Slide 79 text

79 How can we change research? Open tools and infrastructure- to make the last mile shorter Checklists, processes- no excuse that they did not know RSEs – no need to do it all alone Community - to advocate for change @ixek

Slide 80

Slide 80 text

80 @ixek Thank you Contact me: @ixek trallard@bitsandchips.me http://bit.ly/PyConUK-reproducibility

Slide 81

Slide 81 text

81 Credits and special thank yous @ixek ● Chris Holdgraf and all of the Jupyter community https://speakerdeck.com/choldgraf/open-infrastructure-in-the-cloud-with- jupyterhub ● Kirstie Whitaker and the Turing Way community https://doi.org/10.5281/zenodo.3238189 ● Simon Hettrick and the RSE community & Society https://slides.com/simonhettrick/why-recognising-scientific-software- experts-is-key-to-open-science