Non-synomymous terms

• shared: any way of sharing information, could mean I emailed it to you
• publish : citable artifact, discoverable
• archive : long-term preservation

We’ll be focusing on publishing and archiving

• why publish?
• who are we sharing with?
• what materials do we need to publish?
• when do we make them available?
• where do we publish various outputs?
• how do we prepare materials for publication?

For the remaining slides we are going to assume that we are at the point of submitting our manuscript.

• increased visibility / citation

• funding agency (see the NWO Open Science Program)

• journal requirements (see, e.g., PLOS Publishing policies, Nature Publishing policies)

• community expects it

• better research

Reproducibility: what’s in it for me?

more efficient, less redundant science - by allowing others to build upon our work

Five selfish reasons to work reproducibly by Florian Markowetz

• Reason number 1: reproducibility helps to avoid disaster
• Reason number 2: reproducibility makes it easier to write papers
• Reason number 3: reproducibility helps reviewers see it your way
• Reason number 4: reproducibility enables continuity of your work
• Reason number 5: reproducibility helps to build your reputation

Who do we need to share with?

• collaborators
• peer reviewers & journal editors
• broad scientific community
• generally the public

For research to be reproducible, the research products (data, code) need to be publicly available in a form that people can find and understand them. Ideally, both data and code are FAIR.

Activity (in pairs)

Catalog the artifacts you produced during this workshop (3 minutes)

• What needs to be published?
• What does not need to be published?
• Anything that cannot be published?

Advice: One way to determine what you need to publish is to go through and redo the analyses in your paper. Make note of the data and code and notes you needed to do that analysis. Make sure all of that is available. This might seem time consuming, but it assures that what you think you did is what you actually did.

You can make your code and data public at any point of the research process.

However, at the point of paper submission, the results in your paper should be reproducible and therefore the data and code used in the paper published.

• Journals now often require it
• Lets the editor and the reviewers accurately review the paper
• You might want to keep data and code private for just reviewers until the paper is published

• Domain-specific data repository (4TU.ResearchData, EASY | DANS)
• Source code hosting service (GitHub, GitLab, Bitbucket)
• Generic repository (Figshare, Zenodo)
• Institutional repository (e.g., TU Delft Repository)
• Sharing services (RPubs, Dropbox, Google Drive)

Discuss: Research products (code, data) published separately is different from journal supplementary materials. Why?

• is there a domain specific repository?
• is there a plan for long-term preservation?
• can people find your materials?
• is it citable? (does it provide DOIs)
• is your purpose archival, sharing or publication?

Do’s

• non-proprietary file formats
• text file formats (.csv, .txt, .md)

Don’t’s

• proprietary file formats (.xlsx)
• data as PDFs or images
• data in Word documents

Using standard data formats is sometimes required, but even when it’s not, conforming to standards greatly increases opportunities for re-use and understanding.

• top-level README that describes the data or software package
  • list files and naming conventions
  • describe abbreviations, column names, etc.
• installation and usage instructions for software
• citation instructions
  • consider creating a CITATION.cff file which can also be used in R

Documenting your research:

• collect all of the to-be-archived artifacts from the preceding lessons into a directory

• write a README file that describes the contents of the directory

• put a license or waiver on it

Copyright applies to creative works

• source code
• text (manuscripts etc.)
• images

Typically not copyrightable:

• data, results
• individual records in a database of facts

Note: This is not an exhaustive list. Ask your data steward or library if you need help!

• Licenses, copyright, terms of use are complicated issues.
• There are legal implications to your choices.
• Citation is a professional norm in science.
  • We have good systems for ensuring proper citation.
  • Would you try to sue someone in court who fails to cite you properly?
• Keep it simple by putting the least-restrictive license possible

CC0 enables scientists, educators, artists and other creators and owners of copyright- or database-protected content to waive those interests in their works and thereby place them as completely as possible in the public domain, so that others may freely build upon, enhance and reuse the works for any purposes without restriction under copyright or database law.

From the Panton Principles:

[…] in the scholarly research community the act of citation is a commonly held community norm when reusing another community member’s work.

Community norms can be a much more effective way of encouraging positive behaviour, such as citation, than applying licenses. A well functioning community supports its members in their application of norms, whereas licences can only be enforced through court action and thus invite people to ignore them when they are confident that this is unlikely.

Discussion: What are some of the challenges of publishing research products? What are some of the concerns that people have?

More and more specialized staff and services are available at university libraries. They provide great resources for data/software management as well as information and access to repositories. They are particularly good at thinking about data archives and increasingly providing support with code as well.

• Faculty data steward can advise on data management and archival
• Library staff specialized in copyright and licenses
• (Relatively new) University Digital Competence Centres (e.g., the TU Delft DCC) have staff who can help or provide consultance on problems related to research data and software engineering
• They are all very helpful and super awesome!