This otherwise sensible proposition is still mired in 17th century thinking. In any sensible, modern publishing system, "replicated" would be a stage any publication would go through, among, e.g., "peer reviewed", "open data/code verified", "cited", etc.
"Peer-replication model aims to address science’s ‘reproducibility crisis’"
There is no such thing as replication. There is only continued exploration of the reality we live in.
Every experiment is unique.
"Replication" is just a poor word for "Testing if this still happens under slightly different conditions".
Crazy take: There is no "reproducibility crisis". Some fields just have over-interpreted some of their results.
IMO, people spend way too much time thinking of papers as "answers" rather than as one very small piece of a large scientific literature that we are putting together.
@adredish @brembs
Ya for real re: reproducibility crisis. Its a hell of a roundabout way to say "we've constructed a system where we need to grind once idealistic and well-meaning people down to a point where we need to lie to ourselves and each other in an incomprehensible ritual of accumulating tokens without real meaning or cumulative sense of understanding."
It would be rad if instead we had a thing called a scientific literature where we were slowly mapping a terrain with many modalities of thought refracting off a shared basis of observation, but instead we just have an endless churn of vaguely connected PDFs
@adredish @brembs
Like could it be that the prevailing norms in publishing are in fact designed to provide plausible cover for doing work that overclaims as much as possible while surviving just enough scrutiny to land the next grant and improve the JIF for the next quarters shareholder call.
Could it be that the very last thing we should do is pay double to publish results twice to the very organizations who construct the underlying conditions that cause the need for replication in the first place? Its the tail wagging the dog - instead of asking how we get people to do direct replications, we might consider asking why we need to work in such a way that the many many partial replications that happen all the time have no venue, and work is only publishable if it is demonstrably not a replication?
@adredish @brembs
There was a lot of noise made about publishing negative results about a decade ago, but instead of building a system where a continual process of shared results makes a clearer picture, we leaned into the same publication model and the only negative results that are published are big, splashy failures to replicate that are too late to do anything and add little to ongoing practice.
Add comment