About Melavera
Melavera joins two Latin words: mel, honey, and vera, true. Bees gather from thousands of flowers, and the hive turns that work into something that keeps for thousands of years. That's the idea: many careful readers, and one open record that lasts.
Melavera is an open place to read, question and discuss published research from any field and any era. Peer review happens once, behind closed doors, before publication. Here it carries on in public, for as long as a paper matters.
“A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.”
Principles
- Take nobody's word for it. Not a journal's, not a funder's, not ours.
- Funding is always visible. Every paper shows who paid for the research, from what the authors and publisher declared.
- Arguments, not authority. Comments are judged on their reasoning. Credentials aren't required to take part.
- We rank comments, never papers. A vote count can't tell you whether a study is true. Careful reading can.
- Open about how it works. The ranking rules below are the actual rules the code follows.
How comment ranking works
- Two separate votes. You can mark a comment as rigorous (well-reasoned and useful) and, separately, say whether you agree with it. Only rigour affects ranking, so a careful argument you disagree with can still rise to the top.
- Confidence, not raw score. Comments are ordered by the lower bound of the Wilson score interval on their rigour votes. In plain terms: a comment with 3 up and 0 down does not outrank one with 200 up and 30 down. Small samples have to earn their place.
- New comments get a fair hearing. Until a comment has 5 rigour votes, it is "unproven". Every third position in a thread is given to the best unproven comment, so early comments can't lock in the top spots.
- Scores are hidden until you vote. You see a comment's votes only after you've rated its rigour yourself. Visible scores make people follow the crowd; hiding them keeps each judgement independent. A rigour vote can be changed later, but not withdrawn.
- Labelled critiques. Comments can be tagged as methodology, statistics, conflicts of interest, replication or summary, so you can go straight to the kind of scrutiny you care about.
Reading papers here
When a paper is published under an open licence that allows sharing (Creative Commons), you can read the full text right here, credited to its authors. Papers that are free to read but not openly licensed are linked to on the publisher's site, and paywalled papers are linked by their DOI. Either way, the discussion is open.
Melavera never accepts uploaded papers. It stores a paper's public details — title, authors, abstract, funding — from OpenAlex and Crossref, and checks the licence before showing anything more. That keeps the discussion here on solid ground, whoever owns the paper.
Old papers, DOIs and peer review
A DOI (Digital Object Identifier) is a permanent link to a published work. DOIs have only been in wide use since about 2000. Many publishers have since given DOIs to their back catalogues, some reaching back centuries, but plenty of older work still has none.
A DOI says nothing about quality. Preprints, datasets, reports and other unreviewed work get DOIs too. So a paper with a DOI isn't necessarily peer reviewed, and a paper without one isn't necessarily weak.
Peer review as we know it today is also fairly recent. For most of the history of science, editors decided what to print. Nature made sending every paper to outside referees routine only in the 1970s. Einstein's famous 1905 papers were accepted by the editors of Annalen der Physik rather than sent out to anonymous referees, as would happen now. In 1936, when an American journal did send one of his papers to a referee, he was indignant and published it elsewhere.
That's part of why Melavera exists. Review shouldn't be a single gate that a paper passes once. It should carry on, in the open, for as long as the paper matters.
Add a paper to start.