Michael Rowe

Open access licensing

note4 min read

An open licence answers "may I use this?" before anyone asks

Copyright reserves every right to the author by default. An open licence gives some of them away deliberately and in writing — this may be shared, this may be adapted, on these conditions — so that a reader, a teacher, or a machine can reuse the work without negotiating. Creative Commons licences are the ones most scholarly and educational work uses.

Open access licensing

One-sentence definition: Open access licensing means publishing a work under a standard public licence, most often Creative Commons, that grants readers the right to use, share, and sometimes adapt it on stated conditions such as attribution.

Two things are easy to run together here. Open access means a work can be read without paying; open licensing says what you may do with it once you have. A journal article can be free to read and still all-rights-reserved, which means a lecturer can link to it but not put it in a course pack, and a colleague in another country can’t translate it. The Budapest Open Access Initiative (2002), which named the movement, asked for both: free to read and free to reuse. Suber’s Open access (2012) remains the clearest account of why that distinction matters.

The Creative Commons licences are built from four conditions:

  • BY — attribution. Credit the author. Every CC licence except CC01 requires it.
  • SA — share-alike. Anything built on the work must carry the same licence.
  • NC — non-commercial. No commercial use without separate permission.
  • ND — no derivatives. Share it whole; don’t adapt it.

CC BY is the licence most funders and open journals require, because it allows the widest reuse. NC and ND sound protective, but each one cuts off a set of uses the author probably wanted — a textbook publisher, a translation, a remix for a different curriculum — and “commercial” is vague enough that cautious institutions avoid NC material altogether.

What a Creative Commons licence decides for teaching materials

Teaching material is the obvious case. An openly licensed anatomy resource, clinical guideline summary, or assessment rubric can be adapted for a local programme and shared back; a copyrighted one has to be rebuilt from scratch. I license this site under CC BY 4.0 for that reason, and the post on my workflow explains the choice. Knox (2013) is the useful counterweight: openness is not a neutral good, and the assumptions behind the OER2 movement — about who benefits, and what “access” delivers — deserve the scrutiny that paper gives them.

When the reuser is a machine

The licence on a work now also decides whether it can be part of the corpus a language model draws on, and whether a programme can legally give it to an AI tool for its own students. It’s less clear whether the licences work as intended there. Attribution is a condition of every CC BY work, and a model trained on millions of them attributes none of them. The licences were written for people copying documents; the law hasn’t settled what they mean for systems ingesting them.

Sources

Footnotes

  1. CC0 — a public domain dedication, in which the author waives every right they can, so the work can be used for any purpose with no conditions at all, not even credit. Strictly it’s a waiver rather than a licence, which is why it’s the one CC tool without the attribution condition. Wikipedia ↩

  2. Open educational resources (OER) — teaching and learning materials published under an open licence, so anyone can use, adapt, and redistribute them free of charge. A set of clinical skills videos or an OSCE station bank released under CC BY counts as OER; the same material behind a university login does not. Wikipedia ↩

Continue reading

  • Open source software

    Software whose source code is published under a licence that lets anyone read, change, and redistribute it — and what that means for the tools an institution depends on.

  • Publishing with purpose: Using AI to enhance scientific discourse

    The introduction of generative AI into scientific publishing presents both opportunities and risks for the research ecosystem. This essay argues that scientific journals must transform from metrics-driven repositories — prioritising publication volume over meaningful progress — into vibrant knowledge communities using AI to facilitate discourse. AI can support this by surfacing connections between research, making peer review more dialogic, and enabling multimodal knowledge translation. Meaningful change requires coordinated action across institutions, funding bodies, and journals willing to prioritise scientific progress over quantitative metrics. By reimagining journals as AI-supported communities rather than article-processing platforms, the research ecosystem can better serve scientific knowledge development and clinical outcomes.

  • An open scholarship workflow: What academic publishing can learn from open source

    Academic publishing treats scholarship as a finished, individually owned artefact. This post describes a writing and publishing workflow built on a different premise: that a scholarly corpus could work like an open source project — readable, contributable, forkable, and never permanently owned by anyone.