Skip to content

VISUS 2021 supplementary-material recovery checkpoint

GazeForge freezes a reviewed negative-recovery checkpoint for the supplementary event exports described by Barz and Sonntag's 2021 Sensors paper (10.3390/s21124143). This checkpoint extends the VISUS source-resolution record; it does not replace the 2014 benchmark source-authority gate.

Result

No exact supplementary object or authoritative attachment URL was recovered from the reviewed current and archived public surfaces.

This is deliberately narrower than saying that the supplement never existed. The 2021 paper explicitly states that extracted VISUS ground-truth events and predicted events for each scenario and participant were provided as supplementary material. The recovery result therefore records a present evidentiary limitation, not a historical non-existence claim.

The frozen evidence file is:

validation/evidence/visus-source-recheck/
visus-2021-supplement-recovery-exhaustion-evidence-v1.json

Canonical evidence fingerprint:

6e263fe3c8ec202e06cd6c819be622bbe1831202b94393e2bea205df25278761

Exact reconnaissance binding

The checkpoint binds the successful reconnaissance workflow rather than transcribing an unqualified search conclusion:

  • branch: feat/visus-pmc-supplement-probe;
  • exact head: 347ba06dd802f9bd8b38aa2a6c770949702bc3ea;
  • workflow: VISUS PMC supplement discovery, run 34497168884 (run number 7);
  • artifact: visus-pmc-supplement-discovery, artifact ID 10160434617;
  • artifact digest: sha256:4545afdc8a37aee06b8415b68a7bc23614943fa6ac9dcbba774034fed3f7479d.

The record also freezes the raw SHA-256 and internal deterministic fingerprint of all six JSON probe reports in that artifact.

What the reviewed surfaces establish

The exact PMC Open Access package for PMC8235043 contains the article objects and seven figure images but no object classified as supplementary material. The exact PMC XML is 243,122 bytes with SHA-256 2aa36a03effd0ffca21a72cd610a0efe9825d35709c88c20f62db14933b96bc5 and contains PMC metadata whose supplement property is no. That describes the current PMC package; it is not treated as proof that the publisher never hosted a supplement.

Current MDPI article and /s1 requests returned HTTP 403 in the reviewed run. Those are access outcomes, not absence evidence.

A historically observed MDPI static-attachment naming convention was used only to construct a bounded candidate set for sensors-21-04143-s001 across twelve common extensions. All twelve HEAD requests returned 404 and no binary target resolved. Because the filename stem itself was inferred, those responses do not establish the original supplement's name or format.

The corrected Wayback article-namespace queries each returned five ordinary article captures and no supplement candidate. The final static mdpi-res.com/d_attachment/... namespace queries timed out, so the record explicitly keeps complete_negative_search_established=false.

Finally, the archived 2022 MDPI XML replay was successfully retrieved at 324,681 bytes. It contains the target DOI and supplement-related text but exposes zero candidate supplement links. The 2021 archived HTML replay timed out. Accordingly, both archived article captures were not successfully inspected, and no historical attachment URL is promoted.

Why this is a recovery-exhaustion checkpoint

The purpose of the record is to stop repeated blind URL enumeration after the reviewed search reached diminishing scientific value. It preserves the exact negative observations and their limitations so a future recovery attempt can start from a new authoritative lead rather than repeat the same guesses.

The checkpoint therefore keeps all of the following false:

  • supplement object recovered, URL resolved, bytes downloaded, or contents inspected;
  • conclusion that the supplement never existed;
  • claim that every possible public surface has been exhaustively proven negative;
  • full authoritative VISUS source recovery or VISUS dataset licensing;
  • analysis-use or raw-source redistribution rights;
  • participant/stimulus mapping or annotation independence;
  • human-human, model-human, cross-dataset, or native-60-Hz/GP3 validation;
  • Frozen Evidence or any new empirical-performance claim;
  • raw-source redistribution.

The 2021 article's CC BY 4.0 status is not projected onto the missing VISUS benchmark files, and this checkpoint does not resolve reuse terms for an unrecovered supplementary object.

Remaining legitimate recovery routes

Future recovery should proceed only when there is a new authenticated lead:

  1. recover an exact supplement object or explicit attachment URL from an authenticated MDPI, PMC, DOI-linked, or archival publisher surface;
  2. obtain the 2021 supplementary archive or equivalent exact event exports from the paper authors, DFKI, or current VISUS custodians through a normal authorized channel;
  3. use a current author/institutional redeposit that explicitly identifies the 2021 event exports and states the applicable reuse terms.

Any recovered object must be source-authenticated, byte-fingerprinted, and independently reviewed before it can affect source authority, rights, or scientific validation.

Validation

The fail-closed validator is:

from gazeforge.visus_supplement_recovery_exhaustion import (
    validate_visus_supplement_recovery_exhaustion,
)

validate_visus_supplement_recovery_exhaustion(
    "validation/evidence/visus-source-recheck/"
    "visus-2021-supplement-recovery-exhaustion-evidence-v1.json"
)

The validator revalidates the already-frozen 2026-09-09 VISUS authoritative source recheck, binds the exact successful reconnaissance artifact and all six probe identities, and rejects refingerprinted attempts to widen any recovery, rights, source-authority, empirical, or redistribution claim.

No VISUS roadmap checkbox is completed by this negative-recovery checkpoint.