Visible, Trackable, Forkable: Opening the Process of Science
This paper argues that adopting a software-inspired workflow where the scientific process is visible, trackable, and forkable—rather than just its final outputs—would significantly accelerate progress, enhance accessibility and trust, and improve the scalability of quality assurance in science.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you are watching a magician perform a trick. You see the rabbit disappear from the hat, and the magician bows. The audience cheers, and the trick is recorded in a history book as "The Great Vanishing." But you never saw the secret compartment in the hat, the practice sessions where the rabbit fell out, or the times the magician accidentally dropped the rabbit on the floor.
That is how science works today.
The Current Problem: The "Perfect Final Draft" Trap
Right now, scientists work in private for years. They try a hundred ideas, most of which fail. They make mistakes, change their minds, and hit dead ends. But when they finally publish a paper, they only show the audience the "Great Vanishing." They present a polished, perfect story where everything worked the first time.
If a scientist makes a mistake later, the only way to fix it is to tear the whole page out of the history book (a "retraction") and say, "I was wrong." This makes scientists terrified of making mistakes, so they hide them. It also means that if someone else wants to try a different approach, they have to throw away the original work and start from scratch, because they can't see how the original team got there.
The Solution: The "Open Kitchen" Restaurant
The author, Sergey Samsonau, suggests we should run science like a popular open-kitchen restaurant or a collaborative video game, rather than a magic show. He proposes three simple rules to change the "flow" of science:
1. Visible (The Open Kitchen)
The Metaphor: Imagine a restaurant where you can see the chefs cooking. You don't just get the final plate; you can watch them chop the onions, taste the sauce, realize it's too salty, add more sugar, and try again.
In Science: Instead of hiding the messy middle part, scientists would share their "kitchen." You could see the failed experiments, the wrong turns, and the debates. You wouldn't just see the conclusion; you would see the journey. This builds trust because you aren't being sold a magic trick; you are watching the work happen.
2. Trackable (The GPS History)
The Metaphor: Think of a GPS app. If you take a wrong turn, the app doesn't delete your history. It marks the spot: "You went left here at 2:00 PM, realized it was a dead end, and turned right at 2:15 PM." You can see the exact path taken.
In Science: Every single change a scientist makes would be recorded, like a "save point" in a video game. If a scientist changes a number or a method, the system records who changed it, when, and why. If a mistake is found, they don't erase the old version; they just add a new "patch" that fixes it, and the history shows both the error and the fix. This turns "mistakes" from career-ending disasters into normal, fixable parts of the process.
3. Forkable (The Branching Tree)
The Metaphor: Imagine a family tree. If two cousins disagree on a family recipe, they don't have to burn the original cookbook. One cousin can take a copy of the book, change the recipe to add more salt, and keep their version. Both versions exist side-by-side, and you can see exactly where they diverged.
In Science: If two scientists disagree on a theory, they don't have to write two separate, confusing papers. One can "fork" (copy) the other's work right up to the point of disagreement. They can then take their own path. The rest of the world can see both paths, compare them, and see exactly where they differ. It turns competition into a clear, visible map of different ideas.
Why This Matters
- Trust: Instead of trusting a scientist because they have a fancy title, you trust them because you can see their work. It's like trusting a software program because you can read its code, not just because a big company sold it.
- Speed: If everyone can see the "dead ends," other scientists don't have to waste time walking down the same wrong path. They can learn from the mistakes immediately.
- AI and the Future: Today, Artificial Intelligence is getting very good at reading complex data. If science is "open" and "trackable," AI can read the entire history of a discovery and explain it to you in plain English. You could ask an AI, "How did they figure out this cure?" and it would tell you the whole messy, honest story, not just the polished summary.
The Bottom Line
Science has been like a closed club where only the final results are shown. The paper argues that we need to open the doors, show the messy workshop, and let everyone see the tools, the mistakes, and the corrections. Just as software development revolutionized by letting anyone see and improve the code, science could revolutionize by letting anyone see and improve the thinking process.
It's time to stop hiding the magic and start showing the work.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.