jnar — Journal of Negative & Applied Results

About jnar

Methods should get better after they're published — not freeze.

jnar is an open registry built on a simple conviction: the most valuable knowledge about a method shows up after the paper — in the replications, the failures, and the fixes. We give that knowledge a permanent, attributed home.

jnar
Journal of
Negative &
Applied
Results

The name

jnar = Journal of Negative & Applied Results.

The name is the position. Negative results — the things that didn't work — are first-class here, not buried. And "applied" means the practical, what-actually-worked-at-the-bench knowledge that keeps a method reproducible. The cyan n marks where the negative lives.

The reproducibility problem

A method is a living thing. We publish it as a fossil.

When a method is published, it's flattened into a clean, frozen description — a few paragraphs in a supplementary PDF. Everything that makes it actually reproducible is left out: the buffer that had to change, the step that only works cold, the antibody lot that ruined three months of work.

That knowledge doesn't disappear. It moves into people's heads, group chats, and lab notebooks — invisible to everyone outside the room. So the next lab to try the method rediscovers the same dead ends, and the person who originally solved them gets no credit.

jnar's answer is structural: keep the protocol open and versioned, let replications and negative results attach to it as structured, attributed objects, and let the method improve over time with a visible fix history — instead of forking into a hundred slightly-different copies.

How we build

Open, attributed, and built in the open.

Open by default

Methods are public and readable. Knowledge about how science is actually done shouldn't sit behind a paywall.

Credit that compounds

Every version and contribution is ORCID-attributed. When your fix is integrated, you're named — permanently.

Built in the open

We publish our changelog and reasoning. Longevity and trust come from being legible, not from a polished facade.

On who we are: jnar is early, with no users yet and nothing to prove by inflating it. We'd rather show you the model and the protocols than a wall of logos. As real labs adopt it, their work — not our marketing — will be the evidence.

See the model for yourself.

Bring a method in and watch it become a structured, versionable protocol — your first conversion is free.

Try the demo →