Preserve the observation.
Start from an attributed public dataset and disclose any filtering used for display.
THE ANOMALY IS ONLY THE BEGINNING
Keep the
evidence
in view.
A collaborative place to inspect astronomical signals: the observation, the model, what remains unexplained, and the checks that make a claim worth keeping.
Examine a signal ↗Begin with a documented gravitational-wave case from GWOSC.
GW150914OPEN CASE
H1 / OBSERVE
MODEL / TEST
L1 / VERIFY
RESIDUAL
ONE EVENT · TWO DETECTORS · VISIBLE RESIDUALS
01 / SOURCEGWOSC open data
02 / METHODObservation → model → residual
03 / STATUSCurated research demo
01 / SIGNAL WORKBENCH
Open the case.
Every claim should retain the path back to its data. Select a record, change the view, inspect the uncertainty, and write your own reading.
SELECTED RECORD / 01
PRIMARY SOURCE ↗GW150914
The first direct detection of gravitational waves
FILTERED OBSERVATION · H1STRAIN × 10−21
0.250 STIME RELATIVE TO 09:50:45 UTC0.460 S
DATA TREATMENT35–350 Hz band-pass + notch filters
POINTS SHOWNEvery 4th source sample
MODEL NOTENot tuned to erase all signal
INTERPRETATION / PUBLISHED
A signal in two places.
Livingston recorded the event first. Hanford followed about 6.9 ms later. The published comparison aligns and inverts H1 for the detector orientation. A fitted waveform explains much of the filtered trace; the residual still matters.
Read the GWOSC figure notes ↗CROSS-CHECK
01
Independent detector present
H1 + L102
Arrival order
L1 → H103
Published delay
6.9 ms04
Remaining trace
Inspect residualPUBLIC RECORD / EXTERNAL SOURCEWaveform, model, and residual have not been imported for this record. No synthetic trace is shown as observation.
YOUR LOCAL FIELD NOTE
Make a reading. Keep the caveat.
Notes stay in this browser until you export them. They do not alter published findings.
No local note saved yet.
02 / THE METHOD
From noise to a record.
Anomaly research needs visible steps. The route is the product: no conclusion without its observation, model, remainder, and independent check.
Put a model beside it.
Compare an explicit explanation to the trace, without hiding what the explanation misses.
Inspect the residue.
The remainder can contain noise, model mismatch, or a reason to investigate further.
Seek another witness.
Check another detector, method, or instrument before elevating a claim.
03 / OPEN ARCHIVEEvidence deserves
Evidence deserves
an address.
The shared archive is designed to bind a claim to its source, method, dissent, and review history. The current workbench is a local demonstration; public attestation is a planned step.
RECORDOBSERVATIONCROSS-CHECKSOURCE
THE PROPOSED WEB3 LAYER
Residue Credits / RSD
Planned utility: coordinate review bounties, pay for reproducible analysis runs, and attribute contributions to a public case record. No token, contract, rewards, or attestation service is live in this demo.