By Simon Hettrick

Deputy Director, Software Sustainability Institute

Software's role in

open research

25 June 2018 - Open Science day, UCL     @sjh5000      ORCID: 0000-0002-6809-5195

These slides: https://goo.gl/mzxDwb

"scientific knowledge of all kinds should be openly shared as early as practical in the discovery process"

These slides: https://goo.gl/mzxDwb

Experiment

Data

Analysis

Publication

Requires software

Requires software

Requires software

Requires software

If you use software in your research,

your research can't be open until your software is

How much research relies on software?

These slides: https://goo.gl/mzxDwb

Study at: https://goo.gl/gmHwGm

Use

software

Fundamental to

results

Develop own code

69%

92%

56%

44,000 research grants, 7 Research Councils

https://github.com/softwaresaved/software_in_grants_GTR

https://github.com/softwaresaved/software_in_grants_GTR

38%

of total spend invested into software-reliant research in 2017

44,000 research grants, 7 Research Councils

31 repositories

~600,000 papers

https://github.com/softwaresaved/Eprints-searching-for-software

65%

2017

2/3rds

of papers indicate software reliance in 2017

If you use software in your research,

your research can't be open until your software is

If you use software in your research,

Findable

Accessible

Interoperable

Re-usable

These slides: https://goo.gl/mzxDwb

Making software

FAIR

...we're going to need a 9th pillar

It all starts with

a code repository

...and it makes software management so much easier

Findable software

 

  • Easily identified

  • Well described

Identifying software is difficult, giving software an identifier is much easier

 

  • Digital Object Identifier (DOI)

People generally dislike documenting software

 

  • Take an iterative approach

  • Start with a good README

https://gist.github.com/PurpleBooth/109311bb0361f32d87a2

Accessible software

 

  • Stored in a code repository

Interoperable software

 

  • Well described

CodeMeta

https://codemeta.github.io/

Reusable software

 

  • Documented

  • Licensed

Problem

solved!

Lack of awareness

Lack of incentives

Lack of skills

Software Sustainability Institute

Cultivating better, more sustainable research software to enable world-class research

These slides: https://goo.gl/mzxDwb

Software

Training

Community

Policy

29 projects

All 7 Research Councils

Over 4500 learners

137 trainers

80 guides, 50000 views 

RSE campaign

Demographic studies

Finance and career studies

112 Fellows across domains

Over 20 workshops

 

Outreach

20,000+ monthly visits

Vibrant Twitter community

How do you get researchers started with software engineering?

These slides: https://goo.gl/mzxDwb

Command line

Version control

A language

210,000

researchers

23

staff

No. of workshops

Who writes the software used in academia?

These slides: https://goo.gl/mzxDwb

Research

Software

Engineering

Researcher

developer

Research

Software

Engineer

1211

members

29 April 2018

www.rse.ac.uk

@ResearchSoftEng

RSE Groups

RSEs are changing academic software practices

Lack of awareness

Lack of incentives

Lack of skills

Lack of awareness

Lack of incentives

Lack of skills

RSE Conference

3-4 September

https://rse.ac.uk/conf2018/

  • Software is vital to research

  • Open research is not possible without open software

  • We must improve incentives for producing open software

  • We have the skills! Consult a Research Software Engineer

The role of software in open research

These slides: https://goo.gl/mzxDwb

Some easy steps to make your research more open

  1. Get an ORCID for yourself

  2. Use a code repository

  3. Choose an established licence for your software

  4. Keep essential metadata in your repository

  5. Archive major versions of your software and get a Digital Object Identifier (DOI)

  6. Cite your software in your papers (and put your preferred citation in your repository

These slides: https://goo.gl/mzxDwb

Some further reading

  • Good Enough Practices in Scientific Computing: https://arxiv.org/pdf/1609.00037v1.pdf

  • Sharing Detailed Research Data Is Associated with Increased Citation Rate: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0000308

  • Ten Simple Rules for the Open Development of Scientific Software: http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1002802

  • Software citation principles: https://peerj.com/articles/cs-86/

 

These slides: https://goo.gl/mzxDwb

Data:

Source: http://gtr.ukri.org/search/project?term=*

Source: https://twitter.com/sjh5000/status/968060489635901440

Analyis:

Licence: Open Government Licence v2.0

Source: https://github.com/softwaresaved/software_in_grants_GTR

Licence: BSD 3-Clause License

These slides: https://goo.gl/mzxDwb

Contact

s.hettrick@software.ac.uk

@sjh5000

ORCID: 0000-0002-6809-5195

Licence

 © Simon Hettrick 

These slides are licensed under a Creative Commons Attribution 4.0 International 

https://creativecommons.org/licenses/by/4.0/

These slides: https://goo.gl/mzxDwb

Thanks

Neil Chue Hong (for the slide deck)

Software's role in open research

By Simon Hettrick

Software's role in open research

Presentation at Open Science Day, UCL, 24 June 2018.

  • 1,427

More from Simon Hettrick