2  Open Infra

A vital part of open science, but one that is paradoxically easy to overlook, is open infrastructure.

It is easy to overlook because infrastructure (or ‘infra’ for short) is so ubiquitous — the term refers to the tools you use for “doing science”. This includes hardware (such as eye-tracking devices or EEG equipment), software (such as applications to administer online studies or to analyse data), but also conceptual tools (such as tools to develop theory or document and justify decisions).

Open infrastructure is infrastructure that is open source (or more accurately, “free”), which means you have the following freedoms:

In addition, for Open Science, it’s important the infrastructure is community-owned and community-governed, instead of it being governed by a single company, government, or institution.

2.1 How to use Open Infra

Using Open Infra is simultaneously easy and hard. It is easy because it usually only involves switching software (see the next section for examples). It is hard because on the one hand, learning new software can take time and energy; and on the other hand, you may be ahead of your institution’s policies, so officially, maybe only closed science software may be supported. Depending on your situation and on the specific software, you may be able to make the switch (see the next section) or you may have to advocate at your institution (see the section after that).

2.2 Open Infra Substitutions

This list is based on research in health psychology, so it will probably not be applicable to all researchers.

🛠️Task or Function️ 🔠Software type ⛔Closed Science ⭐Open Science
Managing citations or bibliographies Reference Manager EndNote, Mendeley Zotero, JabRef
Writing documents Text Processor Google Docs, MS Word, Apple Pages LibreOffice Writer, Typst, Markdown, Quarto
Collaboration and version control Git Forge GitHub Codeberg, GitLab
Online Studies Data collection Qualtrics, SurveyMonkey LimeSurvey, {formr}
Statistical analyses Data Analysis SPSS, SAS, Stata jamovi, JASP, R & RStudio, Python
Qualitative analyses Data Analysis NVivo, Atlas.ti, Dedoose, MaxQDA ROCK, QualCoder, Taguette, ReQual
Mock-ups and wireframes Design Figma Penpot
Scientific figures Design Adobe Illustrator, Canva InkScape

2.3 Institutions’ responsibilities

However, your institution may not yet have updated their infrastructure to open infrastructure. Implementing new software requires planning and training of staff, as well as potentially overcoming resistance among staff, who can be reluctant to learn new applications (e.g. due to work pressure, a lack of perceived support, various degrees of IT anxiety, aversion to change, etc).

As a consequence, institions can choose to give open infrastructure a low priority. This is ironic, as the longer closed infrastructure is used, the more expensive it will be to transition to open infrastructure later. For example, interoperability, one of the four characteristics of FAIR open data, usually requires open standards, and these are generally implemented in open infrastructure but often missing from closed infrastructure (since propietary software is often developed by corporations with an interest in locking customers into their ecosystems, something made considerably harder by open standards).

As researchers, you can point out the responsibilities of your institutions through a number of channels:

  • Faculties often have some council or committee dedicated to research policy. You can contact your representatives on this council to ask them to bring this forward.
  • Similarly, faculties generally have democratic bodies such as faculty councils. You can contact your representatives here as well.
  • You can also issue a request to your university’s IT department.
  • You can also point to the UNESCO recommendation on Open Science (UNESCO 2021).