Use preregistration for your study plan, and timestamps for everything else you want to date. A preregistration on a registry such as OSF is a time-stamped, read-only copy of your hypotheses and analysis plan, made before you collect or analyse data, and it is what journals and reviewers look for. A file timestamp records a fingerprint of any file, such as raw data, analysis code, lab notes or early ideas, without publishing the file. It shows that exact file existed no later than a certain time. It doesn’t replace a preregistration or a clinical trial registration, and it doesn’t show who had the idea or that the research is sound.
Table of Contents
What preregistration is
The Center for Open Science’s OSF guide defines preregistration as “posting a time-stamped, read-only version of your study plan to a public repository before beginning data collection or analysis”. On OSF:
- a submitted registration is frozen: you can’t edit it or its files, though you can add a documented update or withdraw it, which leaves basic details behind;
- you can make it public straight away or embargo it for up to four years;
- public registrations get a DOI;
- registrations, including attached files, can be up to 5 GB.
OSF accepts study plans, including plans for research already underway or finished if that is disclosed. It does not accept papers, preprints or other research outputs such as data that aren’t part of the plan.
For clinical trials, the rules are stricter. The ICMJE asks for trials to be registered in a public trials registry at or before the first patient is enrolled. A file timestamp is no substitute.
What a file timestamp adds
Preregistration covers the plan. A research project produces many other files that can matter later, in a priority dispute, a misconduct inquiry or a question about whether data was changed:
- Raw data, as soon as it is collected. See how to prove a research dataset existed before a date.
- Analysis code and notebooks, at each key version. See how to prove a Jupyter notebook or result existed before a date.
- Lab notes, protocols and protocol revisions.
- Early ideas and grant drafts, before any plan is ready to register.
- Results and figures before they are shared with collaborators or reviewers.
A timestamp records only the file’s fingerprint, so unpublished data and confidential material stay private. Each file is dated separately, so you can date as many versions as you like.
Side by side
| Preregistration (e.g. OSF) | File timestamp | |
|---|---|---|
| What it covers | Your study plan | Any file |
| Content public? | Yes, now or after an embargo | No, only a fingerprint is recorded |
| Can it be edited? | No, but updates can be added | No; a changed file needs a new timestamp |
| Recognised by journals | Yes, as preregistration | Not as preregistration |
| Shows | The plan existed by the registration date | The exact file existed no later than the recorded time |
| Doesn’t show | That the study followed the plan | Who made the file, or that it is accurate |
How researchers use both
- Before the plan: timestamp early ideas, pilot notes and grant drafts.
- The plan: preregister it on OSF, AsPredicted or a field-specific registry. Clinical trials go on a recognised trials registry.
- During data collection: timestamp raw data batches as they arrive, unchanged.
- Analysis: timestamp code and notebooks before running confirmatory analyses, and again at key changes.
- Before sharing: timestamp results, figures and drafts before sending them to collaborators, reviewers or a preprint server.
This builds a dated trail that sits alongside your preregistration and your lab’s own version control. For the terms people use, see audit trail vs data provenance vs versioning, and for an example dispute, a grant misconduct allegation that turns on one protocol revision.
Where TimeBinder fits
TimeBinder® makes each file’s fingerprint in your browser, and the file stays on your device. The fingerprint is recorded on the Bitcoin blockchain, and you get a Proof of Time Certificate that you or a reviewer can check later. Up to 10 files can go in one TimeBind. A TimeBind shows when a file existed, not who created it or whether the research is right. Read more on blockchain timestamping for academic research.
Frequently Asked Questions
-
Can a timestamp replace preregistration?
No. Journals and reviewers look for a preregistration on a recognised registry. A timestamp is useful for the files a registry doesn’t take, such as data and code.
-
Is an OSF registration timestamped?
Yes. OSF describes preregistration as a time-stamped, read-only version of your study plan. Once submitted, the registration and its files can’t be edited.
-
Can I keep my preregistration private?
On OSF you can embargo a registration for up to four years. A file timestamp keeps the content private for as long as you like, because only the fingerprint is recorded.
-
Should I timestamp raw data?
It can help. A timestamp of the raw data as collected lets you later show that exact dataset existed by a certain time and hasn’t changed since, if it still matches its fingerprint.
-
Does a timestamp prove I had the idea first?
No. It shows a specific file existed by a certain time. It doesn’t show that nobody else had the same idea earlier.
Date your research files
Your first Pooled TimeBind is free, with no credit card required. After that, TimeBinds start from US$5, and an ASAP TimeBind is US$15. See plans and pricing
