We wanted science to become easier to enter, easier to examine, and less dependent on the institution that happened to hold the data.
For me, the first DeSci cycle is a finished chapter. The questions remain open. A movement can lose its momentum while the problem that brought people together becomes more urgent.
My part of that story began with neuroscience: datasets that were difficult to move, methods that were difficult to reconstruct, and useful work that could disappear when a person left a lab. OpSci grew from the possibility that a scientific commons could have infrastructure of its own.
A partial chronology
Start with the work
Brainhack became a place to explore persistent scientific data, git-annex and IPFS. The problem was practical: how could a project remain available and usable beyond one lab’s machine?
Original Brainhack project recordBuild a scientific commons
OpSci brought data sovereignty, distributed resources and open collaboration into the same conversation. The OHBM poster documented that early direction.
OHBM posterFind a language for coordination
DeSci events, shared infrastructure and the Active Entity Ontology for Science asked how scientific communities could coordinate knowledge and action.
AEOS working paperLeave a record others can use
The Filecoin Foundation case study and Brainhack proceedings preserve parts of the commons work and the research into scientists’ needs.
OpSci case studyReturn to the evidence
Glimmer continues the inquiry through research objects, provenance and inspectable workflows. The software is developing; the evaluation still matters.
Glimmer sourceThe acceleration we hoped for
My view is that tools such as ChatGPT and Claude have brought parts of the autonomous-science ambition closer to everyday practice than the first DeSci wave managed. A researcher can now move between an idea, a literature search, code and an analysis with far less friction. In my own work, that changes what is possible for a small team.
But speed leaves several old questions unanswered. Who can access the evidence? Who gets credit? Can a result be independently checked? Can a community sustain the infrastructure it depends on?
These are the reasons I keep returning to Glimmer. An agent’s confident answer is only the beginning of the inquiry. I want to be able to follow a claim back through the objects, choices and executions that produced it.
What I want to keep
The useful inheritance is a commitment to shared knowledge, durable public work and institutions that make participation possible. The names and mechanisms can change. The test is what another person can understand, use and build upon.
This is my working interpretation of a chapter I participated in. The linked chronology records specific artifacts; it does not claim to establish the present health or closure of every DeSci project.
Explore the DeSci record