Help map what we don’t yet know.
OpenFrontiers started as the 12,151 problem records collected for the OpenFrontiers paper. A record of what is open goes stale the moment it is published, so the index is now open: anyone with an account can add a problem, report progress on one, or vote for what should be looked at next.
Pose a problem
Missing from the index? Give a precise statement, its source, and evidence that it is still open. An editor checks it against the existing entries before it is published.
Pose a problem →Submit a solution or progress
Open any problem and choose Submit progress: a full solution, a counterexample, partial progress, a reduction, or a pointer to a paper the entry is missing. Where a problem carries a Lean 4 verification target, a proof of it can be checked in a sandbox — on deployments that have the verifier configured.
Find an open problem →Vote on what matters
Press ▲ on a problem that deserves attention, or on a submission you found useful. Votes order the Most voted view and the list of most-wanted problems below.
See the most voted →Most wanted
No votes yet. Open a problem you care about and press ▲ — the first votes decide what appears here.
How review works
- 01
Recorded
A posed problem waits in the editorial queue. A submission appears on its problem at once, marked Submitted.
- 02
Read by an editor
Editors look for a precise statement, correct attribution, and — for posed problems — an existing entry it duplicates.
- 03
Published or accepted
Publishing a problem earns its author 400 credits; an accepted submission earns 800. Declined work stays on record with the editor's note.
- 04
Status updated
An accepted solution or counterexample can move the problem to partially solved or solved, with the submission linked from the top of the entry.
The full policy is in the editorial guidelines.