In October 2022, the editors of the biology journal eLife published an editorial with a blunt title: Peer review without gatekeeping [1]. They announced that eLife would keep reviewing papers but would eliminate accept or reject decisions after peer review [1]. Every paper it sends out for review would be published with the reviewers’ comments and a short editorial assessment attached [1].

I think this is one of the more sensible experiments in science publishing, because it treats peer review as what it is at its best: expert advice that makes a paper better and helps readers judge it. It does not solve everything. It moves some old problems around, and it leaves a few of the hardest ones, like prestige and hiring, mostly where they were.

What eLife actually changed

It helps to be precise, because the change is easy to overstate. Under the new model, authors first post their paper as a preprint, a public version shared before formal review. eLife editors then decide whether to send it out for review. The editorial is candid that eLife does “not have the capacity to review every preprint submitted to us” [1]. So there is still a selection step.

What disappears is the verdict at the end. In the editors’ words, “we will no longer make accept/reject decisions following peer review” [1]. Every paper that goes to reviewers comes out the other side as a Reviewed Preprint, which contains the manuscript, the individual public reviews, and an eLife assessment [1]. That assessment is meant to summarize the reviewers’ consensus on how significant the findings are and how strong the evidence is, written for a non-expert reader with a common vocabulary [1].

Authors can then choose to revise. The editorial describes revision as an option and never a requirement, and if a revised version goes back to the reviewers, the assessment is rewritten and a new version is posted [1]. This was not a leap from nothing. eLife said it had already posted reviews of more than 2,200 preprints over the previous 18 months [1].

The problems it goes after

The first is speed. Preprint servers already let scientists share results quickly. bioRxiv, described in a 2019 analysis as the largest preprint server focused on biology [2], showed how long the formal route takes: for preprints later published in a journal, the median gap was 166 days, about 5.5 months [2]. Under eLife’s model, reviews appear alongside the preprint, so readers get expert context without waiting for a final yes.

The second is waste. A 2021 study estimated that reviewers worldwide spent over 100 million hours on peer review in 2020 [3]. The authors call their numbers rough estimates, but the scale is hard to ignore [3]. They also point out that “many manuscripts get reviewed at multiple journals, which is a major inefficiency” [3]. When a paper is rejected, those reviews usually vanish from public view and the next journal starts again.

That is roughly the complaint eLife makes about traditional review. Its editors write that the reviews themselves “are rarely seen, their contents reduced to an accept/reject decision” [1]. They also argue that tying review to a publishing decision turns reviewers’ suggestions into requirements, so authors run experiments they think are unnecessary just to get accepted [1]. Take away the final verdict, and a reviewer’s request becomes advice the author can weigh, with readers able to see both sides.

What the new model leaves open

The gate has moved earlier rather than disappeared. Deciding which preprints get reviewed is still a judgment made by editors, and that step is now arguably the most important one. If a paper is never sent for review, it gets no public reviews and no assessment. That decision stays as private as a desk rejection at any other journal.

Readers may also find the new format confusing. A Reviewed Preprint can carry critical reviews and a lukewarm assessment and still sit on the eLife website. Scientists who read the reviews will understand. A journalist, a patient or a student who sees the eLife name on the page may not. The assessment is supposed to help here [1], but it only works if people read it.

Then there is prestige. eLife’s editors argue that journal names have become “the de facto currency of academic research careers” and that scientists are judged by where, rather than what, they publish [1]. I agree with the diagnosis. But one journal changing its process does not change how a busy hiring committee skims a CV. The editorial says it hopes committees will judge the science on their own, while admitting that evaluators do not always have the specialized knowledge to do so and can benefit from the assessments [1]. That is an honest admission, and it shows how much depends on people outside the journal changing their habits. Until the committees change, many scientists will still treat the old prestige signal as the one that counts.

Why I lean toward this model

As a researcher who has been through peer review, I know the traditional system from the author’s side. A lot of careful thinking goes into reviews that most readers never get to see. Some of the most useful comments on a study are its limitations and the questions it leaves open. In the traditional system those comments shape a revision and then stay private.

Making them public treats readers as adults. A paper with a clear account of its weaknesses is more useful to me than one with a journal name and nothing else. It also gives authors some breathing room to argue with a reviewer in the open instead of quietly giving in.

For readers, the practical change is small but real. When you see a Reviewed Preprint, skip the logo and read the assessment and at least one review. For scientists who sit on hiring or grant committees, try doing the same with one candidate’s papers. The model eLife built can only improve how science is judged if the people doing the judging actually use it.

References

[1] M. B. Eisen, A. Akhmanova, T. E. Behrens, J. Diedrichsen, D. M. Harper, M. D. Iordanova, et al., “Peer review without gatekeeping,” eLife, vol. 11, Art. no. e83889, Oct. 2022, doi: 10.7554/eLife.83889.

[2] R. J. Abdill and R. Blekhman, “Tracking the popularity and outcomes of all bioRxiv preprints,” eLife, vol. 8, Art. no. e45133, Apr. 2019, doi: 10.7554/eLife.45133.

[3] B. Aczel, B. Szaszi, and A. O. Holcombe, “A billion-dollar donation: Estimating the cost of researchers’ time spent on peer review,” Research Integrity and Peer Review, vol. 6, no. 1, Art. no. 14, Nov. 2021, doi: 10.1186/s41073-021-00118-2.