{"schema_version":1,"research_id":"eoa_inverse_innovation_exp04_external_evaluation_20260802","source_assessment_id":"layer_decay_and_expiration_management__film_media_production:SENTINEL:v0","cell_id":"layer_decay_and_expiration_management__film_media_production","search_queries":["Frame.io comments version stack comments tied to asset version official","Avid Media Composer review comments sequence version superseded notes official","film editorial review notes expire superseded cut version platform","Frame.io version comments carry over official documentation","video review platform new version comments superseded feedback status resolved notes version official","online proofing comments new version archive previous feedback carry forward official video","film post production review notes version status resolved linked cut workflow","patent media review comments version superseded editing feedback","site:support.filestage.io version comments resolved new version proof comments official","site:help.wipster.io comments versions resolved feedback new version","site:help.autodesk.com ShotGrid notes version status linked official","site:help.syncsketch.com version comments copy forward review notes official","post-production feedback wrong version review rounds film editors","post production supervisor establishes workflows executes notes Producers Guild","DGA Basic Agreement Article 7 director post-production creative decisions director's cut","BLS May 2025 software developers film and video editors hourly wages"],"sources":[{"source_id":"S1","title":"Review Bottlenecks in Post Production: 2026 Fixes","publisher":"Sohonet","url":"https://www.sohonet.com/article/review-bottlenecks-in-post-production-2026-fixes","source_class":"COMMERCIAL_FIRST_PARTY","publication_date":"2026-05-19","accessed_at":"2026-08-02","claims_supported":["Sohonet reports that feedback applied to the wrong version creates unnecessary work and additional review cycles.","Its first-party case account reports review rounds falling from 4.2 to 1.8 across three projects after centralized, frame-accurate review was introduced.","The account is vendor evidence and does not independently estimate the prevalence of stale-note incidents."]},{"source_id":"S2","title":"Improving post-production feedback process","publisher":"Umeå University","url":"https://umu.diva-portal.org/smash/record.jsf?pid=diva2%3A1457277","source_class":"PRIMARY_RESEARCH","publication_date":"2020-08-11","accessed_at":"2026-08-02","claims_supported":["An advanced-level thesis identifies review and feedback on rough and editorial cuts as a complicated, time-consuming part of film and series post-production.","The studied context included online feedback between editors and reviewers and external cooperation with Codemill.","The abstract does not quantify stale-note incidence or the proposed intervention's effect."]},{"source_id":"S3","title":"Basic Agreement of 2020—Article 7: Directors' Minimum Conditions Preparation, Production and Post-Production","publisher":"Directors Guild of America","url":"https://www.dga.org/Contracts/Directors-Economic-Rights/Basic-Agreement-Article-7","source_class":"STANDARD","publication_date":"2020","accessed_at":"2026-08-02","claims_supported":["The agreement requires director consultation on creative decisions in preparation, production, and post-production.","The director is responsible for presenting the director's cut, while the producer or designated person may hold final cutting authority.","These provisions support human authorization of note applicability rather than autonomous software interpretation of creative intent."]},{"source_id":"S4","title":"Producing Credits for Feature Films","publisher":"Producers Guild of America","url":"https://producersguild.org/code-of-credits-feature-films/","source_class":"OFFICIAL_GUIDANCE","publication_date":"2024","accessed_at":"2026-08-02","claims_supported":["Post-production producers provide and implement notes for cuts and oversee the post-production department.","Post-production supervisors establish workflows, manage staff, budgets, and schedules, execute notes, and supervise versioning and delivery.","These roles are identifiable operational adopters or delegated workflow administrators, although the page does not express demand for this particular intervention."]},{"source_id":"S5","title":"Manage different versions of a file","publisher":"Filestage","url":"https://help.filestage.io/en/articles/2211321-manage-different-versions-of-a-file","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"undated; current when accessed","accessed_at":"2026-08-02","claims_supported":["A newly uploaded Filestage version begins without comments, while comments on prior versions remain accessible in file history.","This already prevents prior-version comments from appearing on the new version's active surface by default.","Deleting a version permanently removes its feedback, supporting preservation rather than deletion in an initial test."]},{"source_id":"S6","title":"Copy comments between reviewer groups and versions","publisher":"Filestage","url":"https://help.filestage.io/en/articles/6318439-copy-comments-between-reviewer-groups-and-versions","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"undated; current when accessed","accessed_at":"2026-08-02","claims_supported":["Authorized project-managing team members can selectively copy comments from one version to the next.","Copied comments appear as new unresolved comments authored by the copier, with a new timestamp; mentions are removed.","This substantially overlaps explicit carry-forward but does not document a preserved predecessor-successor identity link or automatic expiry triggered by cut promotion."]},{"source_id":"S7","title":"Webhooks Guide","publisher":"Autodesk","url":"https://help.autodesk.com/view/SGDEV/ENU/?contextId=WE_WEBOOK_GUIDE","source_class":"OFFICIAL_PRODUCT_DOCUMENTATION","publication_date":"undated; current when accessed","accessed_at":"2026-08-02","claims_supported":["Flow Production Tracking webhooks expose entity lifecycle events and support automation of status management.","Its APIs and HTTP event delivery make a revision-triggered lifecycle state machine technically plausible.","Constraints include authentication, six-second delivery responses, possible missing or out-of-order event IDs, seven-day webhook log retention, and customer responsibility for some integration infrastructure."]},{"source_id":"S8","title":"National employment and wage data by occupation, May 2025","publisher":"U.S. Bureau of Labor Statistics","url":"https://www.bls.gov/news.release/ocwage.t01.htm?mod=article_inline","source_class":"GOVERNMENT_OR_REGULATOR","publication_date":"2026-05-15","accessed_at":"2026-08-02","claims_supported":["May 2025 mean and median hourly wages were $71.20 and $65.38 for software developers.","May 2025 mean and median hourly wages were $41.41 and $36.26 for film and video editors.","These are base-labor anchors and exclude contractor premiums, benefits, overhead, specialized post-production rates, software, security, and integration costs."]}],"problem_evidence":{"support":"MODERATE","rationale":"Independent academic work confirms that online rough-cut review and feedback are complicated and time-consuming, while a current vendor account specifically identifies wrong-version feedback, unnecessary work, and extra review rounds. The evidence visibly supports the problem mechanism, but no independent multi-production study quantifies stale-note incidence, attributable rework, or economic loss.","source_ids":["S1","S2"]},"stakeholder_evidence":{"support":"MODERATE","rationale":"The Producers Guild identifies post-production producers and supervisors as workflow owners who administer cuts, notes, schedules, versioning, and delivery; the DGA agreement establishes director consultation and producer or designated-person authority over creative and cutting decisions. These are credible adopters and authorizers, but no named production, studio, or funder expresses willingness to procure or pilot the proposed feature.","source_ids":["S3","S4"]},"prior_art":{"proximity":"SUBSTANTIAL_COLLISION","closest_analogues":[{"name":"Filestage version-isolated comments","similarity":"New versions already start with clean comment surfaces while old comments remain available in version history, providing most of the proposed stale-note demotion and preservation behavior.","remaining_difference":"The documentation does not show automatic expired or superseded states for every prior note, explicit authorized renewal, or per-note predecessor-successor lineage.","source_ids":["S5"]},{"name":"Filestage selective comment carry-forward","similarity":"Authorized team members can selectively copy still-relevant comments to a new version, closely resembling explicit renewal onto a successor cut.","remaining_difference":"The copied item is represented as a new unresolved comment by the copier with a new timestamp and no documented identity link to the predecessor; carry-forward is manual rather than an automatic expiry-and-renewal state machine.","source_ids":["S6"]},{"name":"Autodesk event-driven status automation","similarity":"Flow Production Tracking provides event-driven workflow automation, entity lifecycle events, APIs, and status-management precedents suitable for implementing revision-triggered transitions.","remaining_difference":"The opened guide does not document the complete editorial behavior of expiring every prior-cut note, selectively renewing valid notes, and preserving successor lineage.","source_ids":["S7"]}],"distinctive_claim_remaining":"Relative to version-scoped history with manual or selective carry-forward, promoting a new editorial cut automatically marks every still-active prior-cut note expired or superseded, permits only authorized explicit renewal onto the successor cut, and preserves visible predecessor-successor lineage; this reduces edits attributable to superseded feedback by at least 25% across three productions without increasing total reconciliation and renewal labor.","confidence":"HIGH"},"implementation_evidence":{"support":"MODERATE","rationale":"Version-isolated comment surfaces and selective cross-version copying demonstrate workflow feasibility, while Autodesk documents event-driven status automation and APIs. Implementation still requires reliable cut-promotion events, stable note-to-cut identities, idempotent handling of missing or out-of-order events, role-based authorization, immutable history, audit exports, and retention-policy review. The intervention must demote rather than delete notes and must leave creative applicability decisions with authorized humans.","source_ids":["S3","S4","S5","S6","S7"]},"scores":{"meaningful_impact":{"score":3,"rationale":"Wrong-version feedback can cause unnecessary work and extra review rounds, but independent prevalence and attributable cost are not quantified.","source_ids":["S1","S2"]},"stakeholder_pull":{"score":2,"rationale":"Operational owners and creative authorizers are identifiable, but no named adopter, buyer, or funder has requested or committed to the intervention.","source_ids":["S3","S4"]},"incremental_advantage":{"score":3,"rationale":"Automatic expiry plus lineage could improve consistency over manual resolution, but the advantage remains untested and adds renewal work.","source_ids":["S5","S6"]},"distinctiveness_plausibility":{"score":2,"rationale":"The residual combination is specific, but version isolation and selective carry-forward already implement most of its behavior.","source_ids":["S5","S6","S7"]},"technical_implementability":{"score":4,"rationale":"Existing version models, permissions, copying, lifecycle events, webhooks, and APIs make a prototype credible; event ordering and identity resolution require engineering controls.","source_ids":["S5","S6","S7"]},"adoption_authority_feasibility":{"score":3,"rationale":"Recognized production roles can administer the workflow and authorize creative applicability, but delegation, records policy, and production-specific approval paths must be configured.","source_ids":["S3","S4"]},"evidence_readiness":{"score":3,"rationale":"A bounded retrospective replay is well specified, but it requires proprietary completed-production data and live adjudication by authorized reviewers.","source_ids":["S3","S4"]},"safety_net_benefit":{"score":4,"rationale":"Demotion instead of deletion, preserved history, explicit lineage, read-only replay, and human renewal provide strong reversibility and audit safeguards.","source_ids":["S3","S5","S6"]},"scalability":{"score":3,"rationale":"Event-driven automation can scale across notes and cuts, but identity conventions, permission models, workflow states, and integration infrastructure vary by production.","source_ids":["S4","S7"]}},"score_confidence":"MODERATE","costs":{"first_evidence":{"band_2026_usd":"10K_TO_50K","scope":"Read-only retrospective replay for one completed production, covering at most three sequences, three or more promoted cuts per sequence, no more than 150 notes, a lightweight state-transition prototype, data normalization, and two authorized reviewers' adjudication.","confidence":"MODERATE","assumptions":["Approximately 80-240 software or data-engineering hours and 40-100 editorial, reviewer, and coordination hours.","Completed-production exports are available without material licensing or restoration expense.","The estimate applies contractor and overhead uplift to BLS base-wage anchors."],"source_ids":["S7","S8"]},"initial_deployment_startup":{"band_2026_usd":"50K_TO_250K","scope":"Production-grade integration for one review platform: cut-promotion event capture, lifecycle schema, lineage storage, idempotency, permissions, audit views, migration, testing, security review, and rollback tooling.","confidence":"MODERATE","assumptions":["One existing platform exposes usable APIs or webhooks.","A small engineering and workflow-design team works for roughly three to nine months.","No major media migration, core-platform rewrite, or enterprise procurement fee is included."],"source_ids":["S4","S7","S8"]},"operational_launch":{"band_2026_usd":"50K_TO_250K","scope":"Launch across an initial slate or post-production organization, including workflow configuration, training, support, records-policy review, monitoring, quality assurance, and prospective comparison management.","confidence":"LOW","assumptions":["Launch covers one organization or limited production slate, not a studio-wide global rollout.","Existing identity, review, and audit systems can be reused.","Legal review is limited to workflow and retention configuration rather than litigation or rights remediation."],"source_ids":["S3","S4","S7","S8"]},"annual_recurring":{"band_2026_usd":"50K_TO_250K","scope":"Ongoing integration maintenance, event monitoring, support, training, governance review, audit retention, and periodic accuracy sampling for a limited production slate.","confidence":"LOW","assumptions":["Roughly 0.5-1.5 full-time-equivalent technical and workflow support plus modest hosting.","Media storage and existing review-platform licenses are excluded unless the intervention creates incremental charges.","High-availability, multi-studio, or regulated deployments could exceed this band."],"source_ids":["S4","S7","S8"]}},"verified_pipeline_gates":{"externally_supported_problem":{"status":"YES","reason":"Academic and current industry sources independently establish difficult post-production feedback workflows and wrong-version feedback causing unnecessary work, although prevalence remains unquantified.","source_ids":["S1","S2"]},"externally_credible_adopter_or_authorizer":{"status":"YES","reason":"Post-production producers and supervisors are credible workflow adopters, while directors, producers, employers, or designated cutting authorities are credible authorizers of creative applicability.","source_ids":["S3","S4"]},"distinct_testable_incremental_claim":{"status":"YES","reason":"The remaining claim is narrowly contrastive against version-isolated comments and selective manual carry-forward and has an explicit 25% outcome threshold.","source_ids":["S5","S6","S7"]},"bounded_next_evidence_step":{"status":"YES","reason":"The retrospective replay is limited to one completed production, three sequences, no more than 150 notes, two adjudicators, defined measures, and explicit stop thresholds.","source_ids":["S3","S4"]},"no_unresolved_safety_or_authority_stop":{"status":"YES","reason":"The first step is read-only, preserves source records, makes no autonomous creative judgment, requires authorized adjudicators, and halts on ambiguous identity, disputed transitions, or retention restrictions.","source_ids":["S3","S4","S5"]},"credible_cost_scope_and_range":{"status":"YES","reason":"The four ranges specify bounded engineering and post-production labor scopes and use government wage data as base anchors, while explicitly identifying overhead and integration exclusions.","source_ids":["S7","S8"]}},"next_evidence_step":"With a post-production producer or supervisor and the relevant director, producer, or delegated review authority, replay one completed picture-locked production in read-only mode. Select at most three sequences with at least three promoted cuts each and no more than 150 version-linked notes. Compare the simulated automatic-expiry, authorized-renewal, and lineage workflow with the production's actual version-scoped manual-resolution workflow. Have two authorized reviewers independently classify each transition; measure note-to-cut resolvability, stale-note recall, false expiry, reviewer agreement, renewal minutes, lineage completeness, retrospectively attributable stale-feedback edits, and total reconciliation time. Stop if less than 95% of links resolve, agreement is below 90%, any protected record would be concealed, no material stale-note incidents are found, or renewal burden offsets estimated reconciliation savings. Only then run a prospective controlled comparison across three productions; falsify the efficacy claim if stale-feedback edits do not fall by at least 25% without an increase in total renewal-plus-reconciliation labor.","blocking_evidence":["No independent prevalence estimate quantifies edits or labor attributable specifically to superseded-cut notes.","No named production company, studio, post facility, or funder has expressed willingness, budget, or procurement authority for a pilot.","Historical note exports, cut-promotion logs, and authorized reviewer adjudication require proprietary data and field participation.","The incremental effect over clean per-version comment panels and selective manual carry-forward has not been measured.","Production-specific records-retention, legal-hold, confidentiality, and contractual requirements have not been reviewed.","World novelty, patentability, freedom to operate, market size, and realized impact remain unmeasured."],"research_disposition":"PARTNERED_RESEARCH_PROGRAM","world_novelty_boundary":"This is a bounded ordinary-web assessment. It does not establish world novelty, patentability, freedom to operate, market size, or realized impact; private studio automations, inaccessible products, patents, and undocumented configurations may already implement the remaining combination.","arm":"SENTINEL","candidate_version":0,"controller_recommendation":{"action":"STOP_EMPIRICAL_RESEARCH_NEEDED","repairable":true,"material_progress_observed":true,"progress_targets":["Secure one completed-production partner plus a post-production workflow owner and authorized creative adjudicators.","Obtain cut-promotion logs and version-linked note exports while confirming retention, confidentiality, and legal-hold constraints.","Demonstrate at least 95% note-to-cut resolvability and at least 90% independent reviewer agreement in the bounded replay.","Establish baseline stale-note incidence, attributable edits, reconciliation time, and manual-resolution performance.","Measure false expiry, missed stale notes, renewal burden, lineage completeness, and protected-history visibility.","Proceed to three-production controlled testing and reject the intervention if it fails the 25% stale-feedback-edit reduction threshold or increases total renewal-plus-reconciliation labor."],"reason":"Web research establishes a real but incompletely quantified problem, credible workflow owners, substantial prior-art collision, and technical plausibility. The decisive remaining questions require proprietary production records, authorized human adjudication, and controlled live workflow testing rather than additional bounded web search."}}