Preprints: What Posting One Commits You To

In some fields posting a preprint is standard and expected. In others it is unusual and, in a few, it can complicate publication. The advice therefore cannot be general, and the first step is finding out what your field and your target venue actually do.

What is general is that a preprint is public and effectively permanent, and both of those have consequences worth understanding before posting. For a concrete operational example of how a workplace tool approaches a related coordination problem, see the full explanation.

What it gets you

A public timestamp. Establishes when the work existed, which matters in competitive areas.

Readership before publication, which in fields with long review cycles can be a year or more of visibility.

Feedback earlier, sometimes substantial, and at a stage where it is still usable.

Citability. Others can cite the work while it is under review.

Access. Anyone can read it, including people without institutional subscriptions. This is the argument that matters most to many people and it is a real one.

Evidence for job applications and grants where an output is needed and the paper is in review.

What to check first

What your field does. Ask colleagues, look at what recent papers in your area did. Norms differ sharply and following your own field's is the safe default.

Whether your target venue permits it. Most do now; some do not, and some have conditions. This is stated in the submission policies and it takes five minutes to check.

Whether your co-authors agree. All of them, explicitly. Posting without agreement is a real source of disputes.

Whether the funder requires it. Increasingly some do.

Whether the data or content can be public. Confidential data, embargoed results, anything under a commercial agreement, and anything with identifiable information.

Whether it affects anonymity in review. In double-anonymous review, a preprint can identify you. Some venues accept this explicitly; some do not. Check rather than assume.

What it commits you to

Permanence. Repositories keep versions. You can post a revised version; you generally cannot remove the original.

Which means: post something you are willing to have read as your work. A preprint with an error you later correct still exists in its original form, and the version people cite is frequently not the latest.

Being cited in the preprint version, sometimes for years, including after publication. Update the record where the repository allows.

Public claims. Anything you assert is now on the record, including anything you would have softened after review.

The version to post

Post the version you would send to a journal. Not a rough draft.

Version it clearly, and post updates rather than replacing silently.

Link to the published version when it appears. Most repositories support this and it is frequently forgotten, which leaves people reading and citing a superseded version indefinitely.

Say what stage it is at, on the preprint itself. "Under review at X" or "submitted" is useful information for a reader.

What it does not do

It is not peer review. A preprint has not been evaluated, and the distinction matters when the work reaches people outside the field.

It does not protect against being scooped in the sense people hope. It establishes a timestamp; it does not prevent anyone from working on the same question.

It does not substitute for publication in most evaluation systems, whatever anyone thinks it should do.

The public communication problem

Worth taking seriously, particularly for work with policy or health implications.

Preprints are covered by journalists, and coverage frequently omits that the work has not been reviewed.

If your work is likely to be picked up, consider what happens if it is reported as established. Some authors add a prominent statement on the first page.

Be careful with claims you would soften after review. The preprint version is the one that circulates.

And if you correct something substantive, post the corrected version and say what changed. The correction record matters more for preprints than for journal articles, because there is no other mechanism.

Practical points

Use a recognised repository for your field rather than a personal page. Persistent identifiers, versioning, and it will still be there in ten years.

Choose the licence deliberately, and check it is compatible with your target journal's requirements.

Include the data and code links, if you have them. See sharing data and materials.

Announce it once, wherever your field's conversation happens. Repeated promotion of the same preprint is noticed.

Keep the record straight. Preprint, published version, and any data deposit should all point at each other.

For early-career researchers specifically

Ask your supervisor and co-authors before posting, always. This is one of the more common sources of avoidable conflict.

Understand your field's norm before adopting another field's. Advice from a colleague in a different discipline may not transfer.

A preprint is a genuine output and it can be listed as one, described accurately as a preprint.

The short version

Norms differ sharply by field. Find out what yours does before generalising from anyone's advice, including this.

Check the venue's policy, the anonymity implications, and your co-authors' agreement — in that order and before posting.

It is permanent. Post the version you would send to a journal, not a draft.

Link the published version when it appears, or people will keep citing the superseded one.

For an independent authoritative reference, consult the Open Science Framework for infrastructure for transparent research practice.