2017-18 Dominion

• G-001 SCHEMA EXPRESSIVENESS GAP, four rows. Four set-scope parallel tiers are PROVABLY serial numbered and have a PER-CARD VARIABLE run, which the schema cannot express (EXTRACTION_SPEC section 2.3: print_run and serial_numbered move together, so a variable run collapses to UNKNOWN/UNKNOWN and becomes indistinguishable from 'we found nothing'). What is actually known, enumerated over every rendered row of each tier and not sampled: Main Exhibit Autographs Bronze = SN25 x18 + SN15 x8 over 26 rows; Main Exhibit Legends Autographs Bronze = SN25 x27 + SN15 x3 over 30 rows; Peerless Jersey Autographs Bronze = SN25 x41 + SN15 x15 over 56 rows; Peerless Jersey Autographs Gold = SN10 x55 + SN9 x1 over 56 rows. All four are serial_numbered in fact. The modal token was deliberately not written as a print run. Sources: https://www.tcdb.com/Checklist.cfm/sid/175573/2017-18-Panini-Dominion-Main-Exhibit-Autographs-Bronze ; https://www.tcdb.com/Checklist.cfm/sid/175569/2017-18-Panini-Dominion-Main-Exhibit-Legends-Autographs-Bronze ; https://www.tcdb.com/Checklist.cfm/sid/175675/2017-18-Panini-Dominion-Peerless-Jersey-Autographs-Bronze ; https://www.tcdb.com/Checklist.cfm/sid/175676/2017-18-Panini-Dominion-Peerless-Jersey-Autographs-Gold
• G-002 AN UNHEDGED EDITORIAL PRINT RUN REFUTED BY ENUMERATION, ON THE SAME LINE AND THE SAME SET FAMILY AS THE 2018-19 REFUTATION. cardboardconnection states flatly "PARALLEL CARDS: Bronze #/25 or #/15, Gold #/10, Platinum 1/1 (No Schrempf*)." for Peerless Jersey Autographs - the Gold figure carries no hedge at all. A census of the DISTINCT token set over all 56 rendered rows of tcdb sid 175676 returns SN10 on 55 rows and SN9 on one (PJ-GAN), reproduced on a second differently-parameterised fetch (?sort=0) which returned the same distinct set. Written UNKNOWN/UNKNOWN and value_contested=yes rather than 10. Note the shape is identical to dominion-bkb-2018-19 gap G-002, where the same source's "Peerless Jersey Autographs Silver: 45 cards. Serial numbered #/15." was refuted at SN15 x43 plus SN12 and SN13. Two seasons, one line, one source, the same failure. Sources: https://www.tcdb.com/Checklist.cfm/sid/175676/2017-18-Panini-Dominion-Peerless-Jersey-Autographs-Gold ; https://www.tcdb.com/Checklist.cfm/sid/175676/2017-18-Panini-Dominion-Peerless-Jersey-Autographs-Gold?sort=0 ; https://www.cardboardconnection.com/2017-18-panini-dominion-basketball-nba-cards
• G-004 QUAD MATERIALS COUNT ANOMALY, UNRESOLVED. Three renderings say 39 - tcdb sid 175554 states "Total Cards: 39" and renders 39 rows (numbers 1-40 with 24 absent), cardboardconnection states "39 cards. Serial numbered #/75." and checklistcenter "39 Cards". But this set's own Platinum sid 175557 states "Total Cards: 40" and renders 40 numbers 1-40 with no gap, and its Bronze sid 175555 renders 38 (24 and 33 absent, consistent with cardboardconnection's "Bronze #/25 (No card #33*)"). card_count was written 39 because the independent editorial pair sides with the base page, which is the opposite resolution to the Rookie Showcase case in this same release where the editorial pair sided with the parallels. The extra Platinum row (card 24) is unexplained and no roster was fetched to name it. Sources: https://www.tcdb.com/Checklist.cfm/sid/175554/2017-18-Panini-Dominion-Quad-Materials ; https://www.tcdb.com/Checklist.cfm/sid/175557/2017-18-Panini-Dominion-Quad-Materials-Platinum ; https://www.tcdb.com/Checklist.cfm/sid/175555/2017-18-Panini-Dominion-Quad-Materials-Bronze ; https://www.cardboardconnection.com/2017-18-panini-dominion-basketball-nba-cards ; https://www.checklistcenter.com/2017-18-panini-dominion/
• G-005 NINE SETS CARRY A PER-CARD VARIABLE PRINT RUN ON THEIR OWN ROWS AND sets.csv HAS NOWHERE TO PUT IT. sets.csv has no print_run column (EXTRACTION_SPEC section 2.2 says so and says adding one is out of scope), so the following censuses live only in prose and cannot be queried. Each was enumerated over every rendered row of that set's own tcdb sid, never sampled. Franchise Favorites Dual Signatures (20 rows): SN25 x14, SN10 x4, SN15 x2. Main Exhibit Autographs (26): SN49 x18, SN25 x8. Main Exhibit Legends Autographs (30): SN49 x27, SN25 x3. NBA Champions Dual Signatures (20): SN25 x10, SN15 x4, SN10 x4, SN24 x1, SN4 x1. Peerless Jersey Autographs (57): SN49 x42, SN25 x15. Power Players Autographed Memorabilia (28): SN49 x16, SN15 x7, SN25 x5. Rookie Showcase Jersey Autos (37 rendered): SN49 x33, SN25 x4. Triple Threat Trio Signatures (15): SN25 x7, SN15 x7, SN10 x1. With Authority! Jersey Autos (28): SN49 x18, SN15 x7, SN25 x2, SN24 x1. The corresponding editorial statements all hedge - checklistcenter writes "Serial Numbered #/49 or as Noted" and similar - and a hedge is not a print run. Six further sets ARE uniform and their single figure is recorded in each set's notes: Base Rookies Metal SN199, Rookie Jersey Autographs SN199, Main Exhibit Rookie Autographs SN49, Mammoth Materials SN49, Quad Materials SN75, Quad Rookies Materials SN99, Rookie Dual Signatures SN25, Rookie Quad Signatures SN15. Source: https://www.tcdb.com/Inserts.cfm/sid/171361/2017-18-Panini-Dominion
• G-007 rookie_flag=no was written on all 100 base cards rather than UNKNOWN, and the catalogue is SPLIT on this question. The basis is a within-parent positive discriminative control: the same parent renders the token RC on every one of cards 101-180 and on none of cards 1-100, across four fetches on two endpoints, so the renderer's RC designation is demonstrably live on this page and its absence on 1-100 is a statement rather than a silence. dominion-bkb-2018-19 wrote rookie_flag=no on the same basis; opulence-bkb-2018-19, the closest premium analogue, writes UNKNOWN on all 100 of its base cards. Nothing in either release row tells a consumer which convention it is under. Recorded, not resolved. Sources: https://www.tcdb.com/Checklist.cfm/sid/171361/2017-18-Panini-Dominion?PageIndex=1 ; https://www.tcdb.com/Checklist.cfm/sid/171361/2017-18-Panini-Dominion?PageIndex=2
• G-008 SECTION 5.3 EXCLUSION APPLIED, AND EXISTING RELEASES IN THE CATALOGUE USE THE OPPOSITE CONVENTION. This release sets base_set_size=100 and files cards 101-140 (Base Rookies Metal) and 141-180 (Rookie Jersey Autographs) as separate sets.csv rows, per EXTRACTION_SPEC section 5.3. contenders-bkb-2019-20 sets base_set_size=165 INCLUDING its Rookie Ticket autograph block, and contenders-bkb-2021-22 likewise sets 165; same manufacturer, same evidence shape, opposite convention. dominion-bkb-2018-19 applies the exclusion, so this line is internally consistent across its two seasons. Recorded so the divergence is repairable rather than smoothed over.
• G-009 SET-NAME DIVERGENCE ACROSS THE TWO SEASONS OF THIS LINE, LEFT AS THE SOURCES PRINT IT. The cards 101-140 block is written here as "Base Rookies Metal", which is what both 2017-18 sources print: cardboardconnection heads the section "Base Rookies Metal Set Checklist" (confirmed by a narrow verbatim re-fetch asking for the literal heading strings character for character, alongside "Base Set Checklist", "Rookie Jersey Autographs Set Checklist" and "Franchise Favorites Dual Signatures Set Checklist") and checklistcenter prints "Base Rookies Metal Set". dominion-bkb-2018-19 carries the same structural block as "Rookies Metal", and that season's cardboardconnection page prints "Rookies Metal" without the "Base" prefix. The name was NOT normalised to the sibling's, because names are what sources print; a resolver matching this block across the two seasons will need an alias. Note also that those two 2017-18 sources are plausibly one editorial lineage (CAV-037), so this is arguably one opinion. Sources: https://www.cardboardconnection.com/2017-18-panini-dominion-basketball-nba-cards ; https://www.checklistcenter.com/2017-18-panini-dominion/
• G-011 CAV-036 EXTENDED, FIFTH BEHAVIOUR ON THIS SEASON: THE Inserts.cfm NOTES COLUMN ON THIS PARENT IS ALL IMAGE-COMPLETION PERCENTAGES, ZERO BLANKS, ZERO SERIAL STATEMENTS. One Inserts.cfm fetch on sid 171361 returned all 60 rows (56 under "Insert Sets (56)" and 4 under "Parallel Sets (4)") with no truncation and no interpolated sid, and every single Notes cell reads as a string of the form "7.6% of images added". There are no blank cells, so the both-directions split test cannot be run at all, and there are no serial statements, so the column contributes nothing to any print run. This is the Prizm Mosaic / Contenders shape rather than the prose-versus-bare-token format split CAV-036 was written about; the format trap could not fire here because there is no serial content in any format. Consequence for cost: all 45 print runs in parallels.csv came from a row-level probe on the tier's own sid, one fetch each. Source: https://www.tcdb.com/Inserts.cfm/sid/171361/2017-18-Panini-Dominion
• G-013 ROSTER DISAGREEMENT ON PEERLESS JERSEY AUTOGRAPHS PLATINUM, UNRESOLVED. cardboardconnection writes "PARALLEL CARDS: Bronze #/25 or #/15, Gold #/10, Platinum 1/1 (No Schrempf*).", i.e. it asserts the Platinum roster is short by one. tcdb sid 175677 states "Total Cards: 57" and renders 57 rows including PJ-DSF, matching the base page exactly, while the Bronze (175675) and Gold (175676) sids each render 56 - both lacking PJ-RJK, not PJ-DSF. So the tier that cardboardconnection says is short is the only one tcdb renders complete, and the two tiers tcdb renders short are the two cardboardconnection does not flag. No third source was consulted and the print run itself (1/1) is not in doubt. Sources: https://www.tcdb.com/Checklist.cfm/sid/175677/2017-18-Panini-Dominion-Peerless-Jersey-Autographs-Platinum ; https://www.tcdb.com/Checklist.cfm/sid/175675/2017-18-Panini-Dominion-Peerless-Jersey-Autographs-Bronze ; https://www.tcdb.com/Checklist.cfm/sid/175676/2017-18-Panini-Dominion-Peerless-Jersey-Autographs-Gold ; https://www.cardboardconnection.com/2017-18-panini-dominion-basketball-nba-cards
• G-014 BARE ORDINALS SITTING IN CODED CARD-NUMBER COLUMNS, FLAGGED FOR THE SEPARATE CHASE PASS AND NOT ACTED ON HERE. This pass does not write insert_cards.csv, but the row-level probes surfaced card-number tokens that the next pass must not take at face value. Main Exhibit Rookie Autographs renders a bare "22" as its first row where every other row carries an MR- code, and the identical bare "22" appears on all three of its parallel pages (175577, 175578, 175579). Peerless Jersey Autographs renders bare "25" and "41" ahead of its PJ- codes, on the base page and on all three parallels. Rookie Dual Signatures renders bare "15", "35" and "38" ahead of its RDS- codes, likewise on all four pages. Per CHASE_CHECKLIST_SPEC these are the mixed-code-and-bare-ordinal shape: the real codes settle the scheme and the ordinals should be withheld with the players named. Sources: https://www.tcdb.com/Checklist.cfm/sid/175576/2017-18-Panini-Dominion-Main-Exhibit-Rookie-Autographs ; https://www.tcdb.com/Checklist.cfm/sid/175674/2017-18-Panini-Dominion-Peerless-Jersey-Autographs ; https://www.tcdb.com/Checklist.cfm/sid/175678/2017-18-Panini-Dominion-Rookie-Dual-Signatures
• G-015 TWO PER-ROW FLAGS RECORDED AS EVIDENCE, NOT INTERPRETED. (1) EXCH: Rookie Dual Signatures row RDS-DAL carries the token EXCH on the base page and on all three of its parallel pages - a Panini Rewards redemption in the vocabulary the scope memo documents. It is recorded rather than dropped, because an EXCH row converts an unresolved placeholder into evidenced absence for whoever fills the card-level checklist. (2) UER: Quad Materials card 36 carries the token UER on tcdb's base page for that set. Errors.cfm on the PARENT sid returns "No errors or variations have been added for this set yet.", so the annotation behind that UER was not recoverable and no variation layer was proved or disproved for this release; per CAV-020 that silence is not evidence of absence. No b-suffixed or VAR rows were seen anywhere in the parent enumeration, but Gallery.cfm strips suffixes and Checklist.cfm is lossy, so this release should not be recorded as having no variation set. Sources: https://www.tcdb.com/Checklist.cfm/sid/175678/2017-18-Panini-Dominion-Rookie-Dual-Signatures ; https://www.tcdb.com/Checklist.cfm/sid/175554/2017-18-Panini-Dominion-Quad-Materials ; https://www.tcdb.com/Errors.cfm/sid/171361/2017-18-Panini-Dominion
• G-016 CHANNEL IS UNKNOWN ON ALL 50 PARALLEL ROWS, DELIBERATELY, WITH channel_basis AND channel_confidence LEFT EMPTY. cardboardconnection gives a product-level configuration line ("Product Configuration: 6 cards per pack, 1 pack per box, 6 boxes per case") and nothing in either editorial source, or on any tcdb page fetched, states a trade channel for the product or claims exclusivity for any tier. A product-level configuration is not a channel statement about a tier, and EXTRACTION_SPEC section 2.3 plus migrations/003_parallels.sql require channel_basis to be NULL exactly when channel is UNKNOWN, so unsupported was not written. channel_confidence is likewise empty rather than carrying a value the database does not police (CAV-038). Source: https://www.cardboardconnection.com/2017-18-panini-dominion-basketball-nba-cards
• G-017 SINGLE-LINEAGE EXPOSURE, STATED PLAINLY. cardboardconnection and checklistcenter are the only two editorial sources used here and CAV-037 records that they are NOT interchangeable but remain plausibly one lineage, both descending from Panini's sell sheet. Beckett is unreachable for this season. What rests on that pair plus tcdb: the release date 2018-05-23 (cardboardconnection "Release Date: 05/23/2018" and tcdb "Release Date: May 23, 2018" - two lineages, neither hedged, so this one is the strongest); every sets.csv card_count (tcdb plus both editorial sites agreeing, except Rookie Showcase Jersey Autos where tcdb's base page is outvoted); and the base print run #/75 (cardboardconnection plus the SN75 token on all 100 tcdb rows). What rests on ONE source only: Rookie Quad Signatures Bronze, whose existence and #/10 run come from tcdb sid 182492 alone - cardboardconnection lists only Gold and Platinum for that set and checklistcenter lists no parallels for it. That row is marked source_confidence=single. No marketplace corroboration was attempted for this release, which is a real limit given the scope memo confirmed marketplace evidence works on 2017-18. Sources: https://www.cardboardconnection.com/2017-18-panini-dominion-basketball-nba-cards ; https://www.checklistcenter.com/2017-18-panini-dominion/ ; https://www.tcdb.com/ViewSet.cfm/sid/171361/2017-18-Panini-Dominion ; https://www.tcdb.com/Checklist.cfm/sid/182492/2017-18-Panini-Dominion-Rookie-Quad-Signatures-Bronze
• G-019 PARALLEL ROSTER SIZES ARE NOT UNIFORM WITHIN A SET AND EVERY TIER NEEDED ITS OWN COUNT FETCH. Recorded because it is the planning fact for the next premium 2017-18 build, and because a tier that is short looks identical to a tier that was truncated. Measured, tier roster against its set's card_count: Franchise Favorites Dual Signatures 16/20/20 against 20; Mammoth Materials Gold 16 against 19; NBA Champions Dual Signatures 15/18/20 against 20; Quad Materials 38/18/40 against 39; Triple Threat Trio Signatures 14/15/15 against 15; With Authority! Jersey Autos Tag 24 against 28; Power Players Prime 25 and Tag 24 against 28; Rookie Jersey Autographs Platinum 38 against 39. Two of these are independently corroborated by cardboardconnection footnotes ("Bronze #/10 (No card #9*)" for Triple Threat, "Platinum 1/1 (No DJ Wilson*)" for Rookie Jersey Autographs), which is evidence the short rosters are real rather than truncation. Sources: https://www.tcdb.com/Inserts.cfm/sid/171361/2017-18-Panini-Dominion ; https://www.cardboardconnection.com/2017-18-panini-dominion-basketball-nba-cards
• G-020 CARD-LEVEL CHECKLISTS FOR THE 17 INSERT AND AUTOGRAPH SETS WERE NOT WRITTEN. That is a separate pass governed by CHASE_CHECKLIST_SPEC and this build does not write insert_cards.csv. The row-level probes run here already hold the complete coded roster of every one of those sets at the point they were fetched, so the next pass has an unusually cheap starting point, but nothing was carried into a file. The sids are in each set's notes. Source: https://www.tcdb.com/Inserts.cfm/sid/171361/2017-18-Panini-Dominion

If you know a source that answers one of these, tell us — name the question's label and where to look.