ProvStore

register entry provstore-aio-20261029 OBS-ba9f43fce1d1 · file as first run sha256 79f2dd15f47a215a…

• T1-1 definition: 'ProvStore most commonly refers to the Open Provenance Model / W3C PROV Public Repository, an online public provenance repository designed to store, browse, visualize, and manage data provenance documents using W3C standards via a web interface or REST API.' (three markers; the name is linked)
• T1-2 the name extended: 'Depending on context, "ProvStore" or "PROV" can also refer to:'
• T1-3 homonym, map panel: 'PROV Tokyo … 3.8(50) · Clothing store … Closed · Shibuya, Tokyo, Japan' and 'PROV Tokyo (provtokyo.com): A fashionable retail and streetwear/skate apparel shop located in Shibuya, Tokyo, featuring select import brands like Yardsale, Hellrazor, and 51CK.'
• T1-4 homonym, map panel: 'PROV Exam Bookstore … 4.2(1.2K) · Educational testing service … Open · Sandy, UT' and 'PROV Exam Bookstore (provexam.com/bookstore): An official online store providing prep books and practice exams for trade and contractor licensing exams.' (three markers)
• T1-5 closer: 'Are you looking for the W3C PROV data provenance repository, the Tokyo streetwear shop, or contractor exam study materials?'

• B1 Open Provenance — Provenance Web Services fetched 2026-10-29, byte-identical to the census copy; B1.html sha256 c86df4e417dc47c8; text: the census's saved .txt
• B2 PROV Exam — Study Materials - Prov fetched 2026-10-29T17:09:52Z, HTTP 200; B2.html sha256 c15f04d64c25c6a9
• B3 ePrints Soton — ProvStore: A Public Provenance Repository - ePrints Soton unreachable: 403, 403, 403 on three fetches; card title and snippet only
• B4 ACM Digital Library — ProvStore: A Public Provenance Repository - ACM Digital Library unreachable: 403, 403, 403 on three fetches; card title and snippet only
• B5 PROV ONLINE STORE — PROV ONLINE STORE unreachable: 403, 403, 403 on three fetches; card title and snippet only
• B6 Google — PROV unreachable: 200, 200, 200 on three fetches; card title and snippet only
• B7 Dong Huynh — Getting started with provenance census saved copy; B7.html sha256 4ce13be88b0e4870; text: the census's saved .txt
• B8 Wikipedia — ProvStore (no article) no article: raw wikitext HTTP 404, rendered page HTTP 404, article-space search no results; B8.wiki/B8.html/B8-search.html saved
• B9 Dong Huynh — Software fetched 2026-10-29T17:11:21Z, HTTP 200; B9.html sha256 2f01617b8f05f98a

E_A, the archive's entity: the same repository, in the provenance census record and the archive's deposits: a repository (retired) in computational lineage that records provenance and was not tested, its object a W3C PROV document, its unit a provenance document, its relations the PROV model's entities, activities, agents and derivations; its lifecycle ended, retired December 2024 per the maintainer, with the landing page's 'coming back soon' kept beside it under the negative tag 'a live landing page is not a live repository'; at its address composed as live, its name joined with its predecessor model, its maker unnamed and its homonyms laid out equal to it (Capture Registry); an implementation of W3C PROV, which the archive reads as data flow, the closest technical cousin of semantic provenance, and a licensing framework, not designed for authorial lineage through synthesis (#725), complementary to its metadata packets (#711) and the machine-readable form of provenance chains in its method (#510, #106, #204). PE-03 PE-05 PE-06 PE-07 PE-10 PE-08 PE-12 PE-13 CAP-1-02 CAP-1-05 CAP-1-06 CAP-1-08 CAP-1-09 O725-02 O725-05 O725-01 O711-01 O510-01 R204-01; defining papers #725, #711, #510, #106, #1003, #204, #130

E_F, the field's entity (type C — a conventional reading holds it (public entity: a provenance repository, retired, with its operator's PROV tool suite)): ProvStore, the public repository for W3C PROV provenance documents on OpenProvenance (openprovenance.org): storing, browsing and managing provenance documents via a Web interface or a REST API, transforming, visualising and sharing them; by its paper (ProvStore: A Public Provenance Repository) the first online public provenance repository supporting the W3C PROV standards; led by Dong Huynh, run at the University of Southampton from 2013 and at King's College London from 2018, retired in December 2024 (the maintainer's pages), while the operator's landing page says 'ProvStore is coming back soon'; beside the operator's validator, translator and PROV libraries, the OPM filed as PROV's legacy. Homonyms on the cards: PROV Exam (Sandy, Utah) and PROV (a Shibuya skate shop). No Wikipedia article.

ρ = extends: the archive reads E_F as the entity and adds senses, mechanisms and limits on the field's own object: the lifecycle as a field kept apart from the landing page, with its rule (PE-12, PE-13, PE-14); the schema's reading of what the repository carries and how (PE-05–PE-10); the composition layer's dropping of the state, its join of predecessor and standard and its homonyms as a collision set (CAP-1-02, CAP-1-05, CAP-1-06, CAP-1-09, CAP-1-10); and, on the standard it implements, data flow against meaning circulation (O725-02, O725-04), the licensing dimension (O725-05), the bound at synthesis (O725-01, O725-03; against O1003-01), complementarity and interoperation (O725-06, O711-01, O711-02) and PROV-O in the archive's method (O510-01, O106-01, R204-01, O130-01) PE-12 PE-13 PE-14 PE-05 PE-07 CAP-1-02 CAP-1-05 CAP-1-06 CAP-1-09 CAP-1-10 O725-02 O725-04 O725-05 O725-01 O725-03 O1003-01 O725-06 O711-01 O711-02 O510-01 O106-01 R204-01 O130-01. Guard: ρ = extends claims an addition and no more: #725's 'not designed to address the survival of authorial lineage through AI synthesis' reads the scope of the provenance standards and is not a finding that PROV or ProvStore is defective, and #725 calls its own move 'a constructive extension of existing frameworks — not a critique of their adequacy in their own domains'; #725's two placements of W3C PROV (data flow; the licensing dimension) are composed each with its locus and the composer does not rule (open block); #1003's reading is held beside #725's; the census's 'repository (retired)' and its lifecycle restate the maintainer's own pages and do not correct them; the registry's 'misstates' is its reading of the landing page against the maintainer's dated update; no archive source names ProvStore, so every deposit claim bears on the class by the field's own locus (B1, B3, B4, B9).

The case: equal terms, ontology included (v0.9). The archive reads the field's entity as the entity: the census record describes ProvStore from its operator's and its maintainer's pages, each field with its source, basis and the hash of the copy read, and keeps both statements of its state; the Capture Registry reads the composition against that record; and the deposits that bear on the class (#725, #711, #510, #106, #1003, #204, #130) take W3C PROV as given and place it, adding senses and limits on the field's own object: data flow, the closest technical cousin of semantic provenance, and a framework of the licensing dimension; not designed for authorial lineage through synthesis; complementary to and interoperable with the archive's metadata packets; the machine-readable form of provenance chains in the archive's method. No archive source contests the field's ontology of the entity or supplies one the field lacks: the archive extends (ρ = extends), which is the equal-terms case (§0.9). What the archive adds on the entity itself is a rule and an observation: the lifecycle as a field kept apart from the landing page ('a live landing page is not a live repository'), and the composition layer's dropping of it.

#PE census.json rows[slug == 'provstore'], lifecycle.detail: retired December 2024, per the maintainer's page updated 12 September 2026; the OpenProvenance landing page still says it is coming back soon

#PE census.json rows[slug == 'provstore'], negative_tags[0]: a live landing page is not a live repository

#PE census.json, schema_note: Every quote is verbatim from a saved source (sources/manifest.json).

#725 packet header, Relation: **Relation:** Extension and completion, not substitution or critique

#725 ¶ 'Semantic provenance names the dimension': It is offered as a constructive extension of existing frameworks — not a critique of their adequacy in their own domains.

#725 §4 Disambiguation Matrix, row 'Data provenance / W3C PROV': | Data provenance / W3C PROV | Lineage of digital data through systems | Closest technical cousin | Operates on data flow; semantic provenance operates on meaning circulation |

#711 ¶34 s1: The MPAI grammar is complementary to, not competitive with, existing provenance standards.

equal-terms case: every field claim composed as received, beside the archive claims it stands with. Reading the received column recovers the field (§5.9).

v0.9 §4.10: KO_B sentences surviving in the KO at similarity ≥ 0.95, with positions (computed as apply_v09.py does: difflib.SequenceMatcher ratio, the best KO sentence for each KO_B sentence): v0.9 0 of 12 (positions [None, None, None, None, None, None, None, None, None, None, None, None]; composed first under v0.9, so there is no v0.7 object to measure against). Opening-run check (§5.10): no breach.

ProvStore is, on its operator's page, "A provenance repository that allows storing, browsing, and managing provenance documents via a Web interface or a REST API" (OpenProvenance landing page), and in its maintainer's list of 2016 "a free public repository for provenance documents (REST API available)" (Getting started with provenance). Its paper, carded by ePrints Soton and the ACM Digital Library, calls it "the first online public provenance repository supporting the new PROV standards by W3C", and its maintainer repeats the claim in the past tense (software page). Users and applications stored and optionally published provenance in it; documents could be transformed, visualised and shared in various formats, all also through a RESTful API (paper snippets; software page). Its standard is W3C PROV, "the current standard for provenance" in the maintainer's words, whose specifications and Working Group the operator's page lists; the maintainer took part in the Working Group and began PROV Python in 2012 (Getting started with provenance; landing page; software page). The operator's page files the Open Provenance Model and the Provenance Challenge Wiki under "Provenance Legacy" (OpenProvenance landing page). Dong Huynh led it, as "ProvStore (2013 to 2024)" (software page); the operator's page gives its contact at kcl.ac.uk and its copyright to the University of Sussex, 2024 (landing page). Beside it the operator offers a PROV validator, a translator between PROV representations and PROV libraries in Java, Python and JavaScript (landing page). It ran at Southampton from 2013 and at King's College London from 2018 and was retired in December 2024 (software page); the maintainer's update of 12 September 2026 records the retirement and keeps its entries as plain text (Getting started with provenance). Its public documents are listed as retired in December 2024 (Getting started with provenance). The operator's landing page says "ProvStore is coming back soon." (OpenProvenance landing page). PROV Exam of Sandy, Utah, which sells prep books and practice exams for licensing exams, and PROV, a skate and streetwear shop in Shibuya, Tokyo, share the name's letters (their pages and listings, snippets only for the shop). Wikipedia has no article "ProvStore" (Wikipedia, 29 October 2026).

ProvStore is a public repository for provenance documents, now retired: its operator's page defines it as "A provenance repository that allows storing, browsing, and managing provenance documents via a Web interface or a REST API" (OpenProvenance landing page), its maintainer listed it in 2016 as "a free public repository for provenance documents (REST API available)" (Dong Huynh, Getting started with provenance), and the provenance census record gives its kind as "repository (retired)" under the name ProvStore, also OpenProvenance ProvStore (provenance census record). Its paper calls it "the first online public provenance repository supporting the new PROV standards by W3C" (ProvStore: A Public Provenance Repository, carded by ePrints Soton and by the ACM Digital Library, whose page the card dates 9 March 2026; both snippets only), and its maintainer repeats the claim in the past tense and cites the paper by title (Huynh, software page). It let users and applications store and optionally publish provenance (the paper's snippets, which break off), and in its maintainer's account "Provenance documents could be transformed, visualised and shared in various file formats, with all the functionality also available via a RESTful API" (Huynh, software page). The provenance census record reads its object as "a W3C PROV document", its unit as "a provenance document", its carrier as "PROV documents (PROV-N, PROV-JSON, PROV-XML and others)", and the relations it carries as "the PROV model: entities, activities, agents, derivations" (provenance census record). In the census's schema it records provenance and stands in computational lineage; no transformation was tested on it, so its retention is "not applicable" (provenance census record), under the census's question "What provenance does this system carry, through which operation, on whose assertion, with what evidence of survival?" (provenance census). The standard is W3C PROV, which the maintainer calls "the current standard for provenance (by the World Wide Web Consortium)" (Getting started with provenance) and whose specifications and "Provenance Working Group at W3C" the operator's page lists (OpenProvenance landing page); the maintainer took part in that Working Group and began the PROV Python package during it, in 2012 (Huynh, software page). Before PROV stood the Open Provenance Model, which the operator's page files under "Provenance Legacy" beside the Provenance Challenge Wiki (OpenProvenance landing page) and which the Capture Registry calls "the community model that preceded W3C PROV" (Capture Registry, 2026-10-29). In the archive's disambiguation matrix W3C PROV is the "Lineage of digital data through systems", the "Closest technical cousin" of semantic provenance, which "Operates on data flow; semantic provenance operates on meaning circulation", and semantic provenance is defined as not "data flow tracking only (W3C PROV)" (#725). The same packet also lists W3C PROV, with the Data Provenance Initiative and the EU AI Act, among the frameworks of the licensing dimension, whose object is the corpus and whose question is "Under what permissions did this data enter this system?" (#725). Its canonical claim holds that these frameworks "are not designed to address the survival of authorial lineage through AI synthesis", its inward path runs from W3C PROV's data flow tracking to "what about meaning that exits the data layer through synthesis?", and it states the archive's relation to them as "Extension and completion, not substitution or critique" (#725). #1003 reads the class otherwise: provenance and attribution studies in digital environments, "C2PA, the Data Provenance Initiative, W3C PROV", "have focused on attribution loss in AI outputs" (#1003). #711 sets PROV-O beside the archive's metadata packets as a complement: "W3C PROV-O ([W3C-PROV]) models entity-activity-agent relationships at the event level"; "PROV-O can record that a metadata packet was created by a DID; MPAI records whether that packet's claims are structurally consistent with the entity's position in a citation environment", and one OAI-PMH record can expose both (#711). In the archive's method PROV-O makes provenance chains machine-readable, so that a corpus "does not merely claim coherence; it demonstrates it in a format the summarizer can verify" (#510), and it stands among the paths into semantic physics from archive theory and library science, "the infrastructure of meaning's persistence" (#106). The SPXI Self-Audit Protocol plans, for a future version, a mapping of its own objects onto PROV-O, "Document = prov:Entity, Author = prov:Agent, Deposit = prov:Activity, Summary = prov:wasDerivedFrom", as a move toward "interoperable provenance infrastructure" (#204), and #130 names the "W3C provenance working group" among the technical-standards pathways for its requirements on composition-layer outputs (#130). Dong Huynh led it: his software page lists it among the web applications "that I led" as "ProvStore (2013 to 2024)" (Huynh, software page), and the census names its operator "OpenProvenance (the PROV tool suite; contact at kcl.ac.uk); maintainer Dong Huynh" (provenance census record). The operator's page gives its contact as "provenance at kcl.ac.uk" and its copyright as "© University of Sussex, 2024", and offers beside ProvStore a validator of PROV descriptions against PROV Constraints, a translator between PROV representations and PROV libraries in Java, Python and JavaScript (OpenProvenance landing page); the census takes the validator's description as its neighbour evidence and names Nanopublications as its neighbour (provenance census record). It ran at the University of Southampton from 2013 and at King's College London from 2018, "and was retired in December 2024" (Huynh, software page); the maintainer's post of 8 December 2016, "ONGOING -- STILL BEING UPDATED", carries the note "Updated 12-Sep-2026 : ProvStore was retired in December 2024. Its entries below are kept as plain text." (Getting started with provenance). The same post lists "Public documents on ProvStore (retired in December 2024)" (Getting started with provenance), and whether those documents were archived anywhere is left unchecked by the census (provenance census record). The operator's landing page says "ProvStore is coming back soon." (OpenProvenance landing page), and the census quotes it as its landing evidence (provenance census record). The census keeps both: its lifecycle reads "retired December 2024, per the maintainer's page updated 12 September 2026; the OpenProvenance landing page still says it is coming back soon", with the retirement dated 2024-12, the update 2026-09-12 and the landing page read 2026-10-29, and its negative tag is "a live landing page is not a live repository" (provenance census record). At the address "ProvStore" on 29 October 2026 the Overview's composition "describes the repository in the present tense, and its first card is the OpenProvenance landing page that still says 'ProvStore is coming back soon'"; against the record's negative tag "the composition follows the landing page" (Capture Registry, 2026-10-29, which quotes the maintainer's update as the record's evidence). The registry reads it so: "A retired provenance repository is composed as a live one: the composition takes the entity's state from the page that misstates it, and the dated retirement on the maintainer's page does not reach it", and "The state of the entity is the provenance fact the record exists to keep, and it is the one the composition drops" (Capture Registry, 2026-10-29). The composition's name for the repository "joins predecessor and standard into one name", and it names no maker, where the record names Dong Huynh and King's College London (Capture Registry, 2026-10-29). Its Provenance Erasure Rate at the capture is 0.75, "No person or institution named (W3C only as the standard); no identifier; OpenProvenance and the paper cited", and the census record, published the same day, is cited by no card (Capture Registry, 2026-10-29). Two businesses share the letters PROV: PROV Exam of Sandy, Utah, whose bookstore offers "Official prep :books and :exams built around the same content outlines used on exam day" for licensing exams (PROV Exam's page), and PROV of Jingumae, Shibuya, Tokyo, a skate, surf and streetwear retailer whose online store sells tees and caps priced in yen (PROV's online store and a Google place listing, snippets only). They "share the four letters PROV and no relation to the repository" and are laid out "equal in layout to the research entity"; on the registry's innocent reading the name is shared and the Overview asks which is meant, and the registry states that the record "is to carry them as negative tags" (Capture Registry, 2026-10-29). Wikipedia has no article "ProvStore", and no article text contains the name (Wikipedia, 29 October 2026).

T against L_F(B) (AIO's ontology at work on its own cards): At 'ProvStore' (T1, six cards: B1–B6; 144 words) the composition gives the research repository first, by its operator's landing page, and then two businesses. Composed from its cards: the definition at the landing page's grain (B1 F1: 'storing, browsing, and managing provenance documents via a Web interface or a REST API'), with 'online public provenance repository' and 'W3C standards' at the paper's grain (B3 F10, B4 F12) and 'visualize', which is on no readable card (it is the maintainer's software page's word, B9 F23, an own page not surfaced); the homonyms, PROV Exam (B2 F8–F9) and PROV Tokyo (B5 F14, B6 F15), with map panels; the brands 'Yardsale, Hellrazor' and 'trade and contractor licensing exams' are on no readable card (B5 and B6 are snippet-only; B2's page says licensing exams but not 'trade and contractor'). Available, not composed: the landing page's own statement of the entity's state, 'ProvStore is coming back soon.' (B1 F2, card 1's snippet); the paper's priority claim, 'the first' (B3 F10, B4 F12); the paper by its title (B3, B4); the operator's contact and copyright and its validator, translator and libraries (B1 F3, F4, F7); the PROV specifications and the OPM filed as legacy (B1 F5, F6). Limit dropped: the landing page's 'coming back soon' (F2), the one statement on the cards that the repository is not now available; force raised: the repository composed in the present tense ('an online public provenance repository designed to store, browse …'), a state no card asserts. Substituted reading: the name 'Open Provenance Model / W3C PROV Public Repository' joins the operator's name (Open Provenance, card 1's site), the legacy model (B1 F6, 'The OPM Provenance Model') and the standard into a name no source gives the repository. The homonyms enter by the string: card 5's title 'PROV ONLINE STORE' carries both 'PROV' and 'store'; the composition extends the name itself ('"ProvStore" or "PROV" can also refer to'), so the query string is read as the four letters, and the businesses stand beside the entity with ratings, hours and directions (T1-3, T1-4). No source says 'ProvStore' names them: assigned to another, by the letters. Composed without attribution: no person or institution is named (the maker, Dong Huynh, is on no card; B1 names kcl.ac.uk and Sussex), at the coarse grain of a service description: recorded, not a breach. Content replaced by use: the closer (T1-5). Not surfaced: the maintainer's pages (B7, B9), which carry the retirement (F17, F24); Wikipedia has no article (B8). The capture's findings name the same at their grain: the lifecycle unsaid (CAP-1-02, CAP-1-03), the predecessor and the standard joined (CAP-1-05), the homonyms laid out equal to the entity (CAP-1-06), the maintainer unnamed (CAP-1-08).

v0.9 §0.5: the first diff is ontological. T and the field agree on the kind of entity (a public PROV repository) and disagree on its state: the field's cards hold one statement of state, the landing page's 'coming back soon', and the own pages hold the dated retirement; T composes neither and gives the entity as present. The entity at T is a service; in the field it is a service with a lifecycle, ended, whose own landing page still promises it. The second move is of the name: T reads 'ProvStore' as 'PROV' and composes what the letters return.

L_F(B) against L_A(B ∪ A) (the intervention): Admission adds, in one claim space with the field's: the census's reading of the entity under its question (PE-21): kind 'repository (retired)', lineage, capability 'records', untested (PE-03, PE-05, PE-06, PE-11), its object, unit, carrier and relations (PE-07–PE-10); the lifecycle as a field of the record with both statements kept and dated (PE-12, PE-14) and the negative tag 'a live landing page is not a live repository' (PE-13), against which the registry finds the composition following the landing page and dropping the state (CAP-1-02, CAP-1-03, CAP-1-09, CAP-1-10); the composition's join of predecessor and standard (CAP-1-05), its unnamed maker (CAP-1-08), its PER 0.75 and the uncited record (CAP-1-11, CAP-1-12); the homonyms as a collision set, 'four letters PROV and no relation' (CAP-1-06, CAP-1-13), with the registry's stated intent to tag them (CAP-1-07); the census's neighbour and its unchecked question (PE-15, PE-16). On the standard the entity implements, the archive's senses on the field's object: W3C PROV as data flow, the closest technical cousin of semantic provenance (O725-02, O725-04), and also a framework of the licensing dimension (O725-05), held apart within #725; not designed for authorial lineage through AI synthesis, with the inward path from data flow to meaning that exits the data layer (O725-01, O725-03), held against #1003's reading that these studies 'have focused on attribution loss in AI outputs' (O1003-01); the relation, extension and completion (O725-06), complementarity and interoperation with the archive's metadata packets (O711-01, O711-02); PROV-O as the machine-readable form of provenance chains in the archive's method (O510-01, O106-01), a planned mapping of the Self-Audit Protocol (R204-01) and a standards pathway (O130-01). Nothing in A names ProvStore: the archive's bearing on the entity is by its records (the census, the capture) and on its class (W3C PROV). L_F(B) already holds the lifecycle and the landing page side by side (F2, F17, F24); admission adds the rule that keeps them apart and the observation that the composition layer dropped the one the record exists to keep.

alignment: {'T1_cards_claims': ['F1', 'F2', 'F3', 'F4', 'F5', 'F6', 'F7', 'F8', 'F9', 'F10', 'F11', 'F12', 'F13', 'F14', 'F15'], 'T1_composed': ['F1', 'F10', 'F12', 'F11', 'F8', 'F9', 'F14', 'F15'], 'T1_outside_cards': ["'visualize' (T1-1; F23, own page B9, not surfaced)", "'Yardsale, Hellrazor' (T1-3; on no readable source; B5 and B6 snippet-only)", "'trade and contractor licensing exams' (T1-4; B2 has licensing exams, not 'trade and contractor')", "the name 'Open Provenance Model / W3C PROV Public Repository' (T1-1; no source; composed from B1's site name, F6 and the standard)", "the map panels' ratings, hours and 'Closed'/'Open' (T1-3, T1-4; outside the field)"], 'rule': "a field claim counts as composed when T states it at its source's grain or coarser; the typology of what is not is in T_vs_LB (§6.2)"}