Open science and commoning beyond licensing

Deteriorating digital commons in the absence of collective governance

Bastian Greshake Tzovaras

📫 bgreshake@proton.me
🌐 tzovar.as
🐘 @gedankenstuecke@scholar.social


right

2026-06-10
In Academia

Positionality

  • started in biology, ecology & evolution, …
  • moved into bioinformatics, …
  • and from there into citizen science, CBPP

In open knowledge/science/source/culture

Positionality

  • co-ran openSNP 2011-2025
  • regular contributor to Wiki* projects, OpenStreetMap
  • Volunteers with CoMaps & Codeberg
  • 'AI' hater1
  • this talk draws from anecdata

What are our commons?

Examples of commons in open science

  • Free & Open Source Software tools: pyOpenSci, rOpenSci, numpy, …
  • Online Platforms: Zenodo, Open Science Framework, Rogue Scholar…
  • Citizen Science projects/platforms: Zooniverse, iNaturalist, …
What are our commons?

What makes a digital commons?

Commonly 🥁,

  • some form of 'open' license
  • some form of crowdsourcing/involvement of people to create the common output
  • c.f. involvement in citizen science
Image: CC BY-SA, Katharina Kloppenborg
A narrow view of the commons

The problem

  • focus on output of a commons, not the process of creating it
  • outputs: knowledge or product (software, data, …)
  • shortsighted: for many (most?) commons the 'value' lies in the community maintaining it/continuing to create it
    • example: Wikipedia ./. Grokipedia
Including the human factor

Governing a commons

From Ostrom's design principles:

  1. Collective-choice arrangements: Most individuals affected by the operational
    rules can participate in modifying the operational rules
Ostrom E. Analyzing Long-enduring, Self-organized, and Self-governed CPRs. In: Governing the Commons: The Evolution of Institutions for Collective Action. Canto Classics. Cambridge University Press; 2015:58-102. | Image: CC BY-SA 3.0 by Holger Motzkau
Including the human factor

Levels of governance

Types of involvement in Ostrom's Institutional Analysis and Development framework

  1. Operational Choice: ~ what can an individual contributor do?
  2. Collective Choice: rules for what can & cannot happen on the individual level
  3. Constitutional Choice: how are the procedures for collective choice defined/changed
McGinnis, M. D. (2011). An introduction to IAD and the language of the Ostrom workshop: a simple guide to a complex framework. Policy Studies Journal, 39(1), 169-183. (revised 9/21/2020)
A narrow view of the commons

The problem (part 2)

  • Most open science/knowledge commons do not implement collective choice mechanisms, let alone constitutional choice
    • Result of focusing on outputs, not social dynamics
  • Leads to conflicts as time passes, with no mechanisms for resolution
If you don't like it - fork it

Examples from FLOSS

  • informal collective choice clashes with benevolent dictator for life/corporate constitution
  • just fork it ignores the very real cost of doing so
    • financial
    • organisational / infrastructural
    • social
forgejo is the software that powers Codeberg's code forge
An example from open/citizen science

iNaturalist & generative 'AI'

  • organisation taken by surprise about community reaction
  • implemented project despite criticism
  • neither collective nor constitutional influence for community
Wikimedia Foundation & Wikimedians

Wiki(m|p)edia

  • strong collective choice for contributors, community rules & norms shaped by contributors
  • some constitutional choice: 50% of WMF board are nominated by community/affiliates
Sharing power with the commoners

Degrees of participation

  • For many of our commons: somewhere between informing & consultation
  • Even in more extreme cases, constitutional choice is limited
  • These examples are not rare: open knowledge/science commons are often run by self-perpetuating boards:
    • Open Source Initiative, Mozilla, Center for Open Science, …
FAFO

Consequences

  • when (not if) things go wrong, the contributors can only complain and vote with their feet
    • best case: a clear winning fork but lots of resources are spent to create it
    • worst case: both commons wither, ultimately leaving both dead
  • over time commons become brittle through accumulated governance debt
Welcome to Vereinsmeierei

What alternatives are in-use?

  • Moving away from benevolent dictator for life models (c.f. Python & Mastodon)
  • Codeberg, the open source-focused GitHub alternative, is run as a member-controlled association (e. V.) out of Germany
  • OpenStreetMap is hosted under a member-run non-profit out of the UK
  • Rogue Scholar will become a German member-based non-profit in 2026
Costs of switching

Why do we see few alternatives?

  • Many commons grow 'unplanned' (scratching an itch)
    • creates a 'local governance maxima'
  • Changing the model is hard and often not a priority
  • Very little guidance easily available, cf. choosealicense.com etc.
source of Haggis unknown, tell me if you know which book it's from!
The exception proves the rule

Do things always go wrong?

  • openSNP was run by a group of 3 BDFLs
  • Open Humans used a community seat model like Wikimedia
  • Lack of conflict because?
    • 'Data donation' as transactional model?
    • Perceived value for most use cases of these commons is in the data, not community
What to do?

Considerations for the future

  • A call to collectively look beyond the outputs of a commons (especially those built on collaboration)1
  • If you are (co-)running a commons: please consider community governance processes before $@!# hits the fan
  • If you are contributing to a commons: the second best time to start lobbying for joint decision making power is now
1 but even for 'pure' collection infrastructure, not having a collaborative governance is problematic, c.f. the whole FLOSS ecosystem depending on Microsoft's GitHub
Thanks! 💖

Thanks for your attention!

You can