{
 "question": "Round 2 gap-filling research for a deep-time human-knowledge preservation project (Memory Commons, memorycommons.foundation). Round 1 already verified: GitHub Arctic Code Vault (21TB/186 piqlFilm reels, rule-based selection), Rosetta Disk (13k pages/1500 languages, optical-only decoding), EchoStar XVI \"The Last Pictures\" (GEO artifact with hydrogen-line/pulsar primer), FCC 5-year LEO deorbit rule (2022), GEO/graveyard ~perpetual persistence bounded by ~7.6 Gyr solar red giant, lunar abrasion rates (0.2 mm/Myr cm-scale vs 20 mm/Myr 10m-scale), 5D silica projections (360TB projected vs 3 bits/spot demonstrated 2013), tungsten/Si3N4 Arrhenius gigayear media (1.8 eV barrier), Arch Mission Nanofiche (2000 pages/cm\u00b2 at 150dpi, vendor-projected longevity), DNA storage promise vs economics, MIAB two-tier Golden Record successor (Jiang 2023). DO NOT re-research those. Research these SPECIFIC GAPS with numbers and citations: (A) Complete the preservation-attempts survey: Voyager Golden Records (construction, contents, trajectory, expected survival ~billions of years, Oberg/Space.com longevity estimates); Pioneer 10/11 plaques; LAGEOS-1 plaque and its ~8.4 Myr stable 5,900 km orbit; KEO satellite project failure history (repeated delays 2003-present, current status); Memory of Mankind Hallstatt salt mine (ceramic microfilm tablets, token-map funding model, Martin Kunze); Sanctuary/Arnano sapphire disks on the Moon (Chang'e 6? status) and MoonArk (Carnegie Mellon, Peregrine); Beresheet crash 2019 \u2014 did the Arch Mission Lunar Library survive impact (foundation's assessment); Astrobotic Peregrine January 2024 failure and loss of payloads (burned in reentry); Lunar Codex (Samuel Peralta, NASA CLPS); Celestis and LifeShip current products AND EXACT PRICING tiers; Microsoft Project Silica current status (Azure, 7TB glass, 10,000 yr claim); Arctic World Archive / Piql commercial model; Svalbard Global Seed Vault funding structure (Norway government + Crop Trust). (B) Safe-orbit physics expansion: graveyard orbit (GEO+~300km) long-term stability studies; GEO inclination oscillation (~15\u00b0 amplitude, ~53-year luni-solar cycle); MEO disposal orbit instability via lunisolar resonances; LAGEOS-type stable MEO shells; Lagrange point L4/L5 long-term stability for artifacts; lunar lava tubes and permanently shadowed craters as archive vaults; micrometeoroid flux at GEO vs LEO (grams/m\u00b2/year); radiation darkening of fused silica in space. (C) Materials matrix with COST: Cerabyte ceramic-on-glass storage (density, cost projections, funding); Arnano sapphire disc specs (200GB? analog etching, readable with microscope); Project Silica cost/GB if published; nickel electroforming/NanoRosetta mastering costs (USD per master); synthetic DNA storage cost per MB 2025-2026; piqlFilm specs (capacity/reel, longevity claim, cost). (D) Rideshare/hosted-payload economics: SpaceX Transporter rideshare pricing per kg (current official price list); GEO hosted payload cost ranges (per kg or per slot, Astranis/SES/Intelsat programs); lunar delivery cost per kg (Astrobotic ~$1.2M/kg?, Intuitive Machines, Firefly CLPS prices); cubesat build+launch total cost ranges; payload qualification requirements (NASA GEVS vibration, outgassing ASTM E595) and typical certification cost; DHL MoonBox and other small-keepsake lunar delivery retail pricing. (E) Funding models VERIFIED: Celestis pricing tiers and revenue model; LifeShip $ pricing; Long Now Foundation membership/donation model and annual budget; Internet Archive annual budget and funding mix; Wikimedia Foundation budget/funding mix; Software Heritage funding (UNESCO partnership, sponsors) \u2014 as the closest living \"knowledge commons with institutional backing\" precedent; Memory of Mankind token pricing. (F) Provenance/integrity/governance for century-scale archives: C2PA content credentials adoption status 2025-2026; RFC 3161 timestamping and long-term signature validation (ETSI LTV/AdES, evidence records RFC 4998); post-quantum signature standards (NIST FIPS 204/205, SPHINCS+ for long-lived signatures); trusted timestamping via transparency logs (certificate-transparency-style, Sigstore/Rekor) for archival integrity; Ostrom's 8 commons-governance principles applied to digital commons; Wikimedia/Internet Archive/Software Heritage governance structures as anti-capture models; examples of archive capture or content disputes (what failure looks like). Output: cited, number-heavy findings per topic with confidence levels; flag vendor claims vs peer-reviewed vs regulatory sources.",
 "summary": "Round 2 filled major gaps in sections A-E with 24 verified claims, but coverage is uneven. Safe-orbit physics (B) is now well-grounded: peer-reviewed GEO graveyard stability evidence exists only at ~200-year horizons with initial perigee and eccentricity as the controlling parameters, MEO disposal orbits are chaotically unstable due to overlapping lunisolar resonances (all Galileo graveyard-definition attempts failed), GEO derelicts near the 165E/15W unstable equilibria are effectively unpredictable, and lunar lava-tube caves offer ~290 K near-isothermal (<1 C diurnal swing), radiation- and micrometeoroid-shielded vault environments. The materials matrix (C) gained hard numbers: nickel (Lunar Library = 100 g, 25x40 um discs; NanoRosetta 81,000 pages per 160 mm wafer, glass-master-to-electroform process, no published master cost), ceramic microfilm (5M chars per 20x20 cm MOM tablet), piqlFilm (120 GB/reel, 1000+ yr vendor claim), Project Silica (7 TB raw per 120 mm platter, TRL 6), Cerabyte (125 GB now to 1 TB+ per medium, <$1/TB end-of-decade target), and DNA (>$100k/GB write, >$500/GB read as of 2022 vs IARPA's $1/GB 2025 goal). Economics (D/E) is anchored by SpaceX's official rideshare list ($350k for 50 kg to SSO, $7k/kg in 2026, escalating ~$500/kg/yr to $7.5k/kg in 2027) and MOM's EUR 60-350 retail tablet-plus-token funding model; the Arch Mission Lunar Library's survival of the 2019 Beresheet crash remains a foundation self-assessment, not NASA-confirmed. Section F (provenance/integrity/governance) and much of the requested survey (Voyager, Pioneer, LAGEOS, KEO, Celestis/LifeShip pricing, institutional archive budgets) produced no surviving verified claims and need a dedicated follow-up round.",
 "findings": [
  {
   "claim": "Arch Mission Lunar Library (Beresheet, April 2019): the foundation's official position is that the Library survived the crash and is intact on the Moon, based on its own scientific advisors' reading of NASA LRO imagery \u2014 a self-assessment, not independent or NASA confirmation (NASA's LRO team saw only a ~10 m dark smudge and could not confirm an intact lander). The artifact itself is a 100-gram, 120mm-DVD-sized stack of 25 nickel discs, each 40 microns thick, manufactured by NanoArchival (Bruce Ha / Stamper Technology); mass math checks out (~11.3 cm^3 x 8.9 g/cm^3 Ni ~ 100 g).",
   "confidence": "high",
   "sources": [
    "https://www.archmission.org/spaceil"
   ],
   "evidence": "Primary page verbatim: 'Currently it is believed that the Lunar Library survived the crash of Beresheet and is intact on the Moon according to our team of scientific advisors based on imagery data provided by NASA's LRO' and '100 gram nanotechnology device that resembles a 120mm DVD... 25 nickel discs, each only 40 microns thick... made... by NanoArchival.' Specs corroborated by Blocks and Files (2023), PRWeb 2019 release, Fast Company; survival NOT corroborated by NASA/ASU LROC team, which matches the claim's self-assessment framing.",
   "vote": "merged [0]+[1], both 3-0"
  },
  {
   "claim": "Memory of Mankind (Hallstatt, Austria) archive design: a 20x20 cm ceramic microfilm tablet carries up to 5 million characters (= five 400-page books) at 5 lines/mm, readable with a 10x magnifier; a book needs 1/200 of its printed volume. The archive sits 2 km inside the salt deposit of mountain Plassen and self-seals because salt creeps ~2 cm/year into voids; MOM claims pressure on tablets (even under a hypothetical 5 km ice shield) stays ~1/5 of material burst pressure. All figures are vendor/project self-reported, though the salt-creep rate matches published salt-mine (WIPP-type) behavior.",
   "confidence": "high",
   "sources": [
    "https://www.memory-of-mankind.com/how-is-information-kept-legible-for-1-million-years/",
    "https://www.memory-of-mankind.com/"
   ],
   "evidence": "Primary page verbatim: '5 million characters, this equals 5x400-pages books'; 'five lines per mm... easily readable with a 10x magnifier'; '1/200 of the volume'; 'salt flows with a speed of 2 cm/year into any void'; pressure 'approximately a fifth of the burst pressure of the used materials.' Corroborated by American Ceramic Society (ceramics.org, 2023) and Wikipedia. Internal math consistent (2,500 chars/page). Note: '2 km deep' almost certainly means tunnel distance, not vertical overburden (Plassen summit is 1,953 m); pressure figure has no published engineering analysis.",
   "vote": "merged [2]+[3], both 3-0"
  },
  {
   "claim": "MOM funding model (section E): direct retail sale of tablet space \u2014 EUR 60 for 1/4 tablet, EUR 150 for a full tablet, EUR 350 for a full tablet plus a physical duplicate in a gift box; every purchase includes a ceramic token (fired at 1,200 C, embossed with a map to the Hallstatt archive entrance) mailed to the buyer; gift vouchers span EUR 60-350. Retail sales are the primary but not sole channel (donations also accepted, incl. a 'A Thousand Books' campaign with EUR 500/1,000 tiers).",
   "confidence": "high",
   "sources": [
    "https://www.memory-of-mankind.com/purchase-a-mom-tablet/",
    "https://www.memory-of-mankind.com/product/1-4-tablet-in-the-mom-archiv-plus-token-by-mail/"
   ],
   "evidence": "Live shop pages fetched 2026-07-05 confirm exact prices: EUR 60.00 (1/4 tablet), EUR 150.00 (1/1 tablet), EUR 350.00 (1/1 + duplicate, free shipping); voucher listed at '\u20ac60.00 \u2013 \u20ac350.00'; all products include 'the token by postal service.' Token material (ceramic, map to entrance) corroborated by Wikipedia, ceramics.org, ceramicartsnetwork.org. Note: the site never uses the word 'Premium' for the EUR 350 tier.",
   "vote": "3-0 [4]"
  },
  {
   "claim": "MEO disposal/graveyard orbits are NOT indefinitely stable parking places: the GNSS MEO region (incl. Galileo, ~29,600 km SMA, 56 deg inclination) is permeated by overlapping lunisolar secular resonances producing chaotic, diffusive motion, which is why all past attempts to define stable Galileo graveyard orbits failed \u2014 the region 'is far too complex to allow for an adoption of the simple geosynchronous disposal strategy.'",
   "confidence": "high",
   "sources": [
    "https://academic.oup.com/mnras/article/464/4/4063/2527864"
   ],
   "evidence": "Rosengren et al., MNRAS 464(4):4063-4076 (2017), abstract verbatim: MEO 'permeated by a devious network of lunisolar secular resonances, which can interact to produce chaotic and diffusive motions' and 'all past efforts to define stable graveyard orbits, especially in the case of Galileo, were bound to fail.' Corroborated by Rosengren 2015 (MNRAS 449), Daquin 2016 (CeMDA), Chaos 29:101106 (2019). Gondelach/Armellin 2019 shows some disposal orbits are practically robust over ~200 yr, which qualifies century-scale engineering but not indefinite stability. NOTE: the specific '~55-year Lyapunov time' figure was REFUTED in verification (1-2) \u2014 do not cite it.",
   "vote": "3-0 [5]"
  },
  {
   "claim": "GEO graveyard (super-synchronous) orbit stability: in 200-year parametric propagations with Southampton's DAMAGE propagator (Lewis, Swinerd, Martin & Campbell, Acta Astronautica 55:299-310, 2004 \u2014 not 'Chao & Baker'), initial perigee altitude and initial eccentricity are the dominant factors for keeping a disposed object above the GEO protected region. Critically for deep-time claims: peer-reviewed graveyard stability evidence exists at century scale (~120-1,000 yr horizons across the literature) with force models of low-order harmonics + luni-solar third body + SRP; NO peer-reviewed gigayear-scale GEO propagation exists \u2014 the 'billions of years / until the Sun goes red giant' persistence figure traces to Trevor Paglen's Last Pictures publicity, not peer review.",
   "confidence": "high",
   "sources": [
    "https://www.sciencedirect.com/science/article/abs/pii/S0094576504001924",
    "https://eprints.soton.ac.uk/22758"
   ],
   "evidence": "Abstract verbatim: 'initial perigee and eccentricity of the disposal orbit are the most important factors for maintaining the orbit above the protected region' over 'a 200-year period' with 'low-order gravitational harmonics, third-body influences and solar radiation pressure.' Adversarial search found peer-reviewed horizons clustering at 120-1,000 yr (Gachet 2019 CeMDA; Muciaccia 2021); the gigayear figure appears only in Space.com/Fast Company 2012 coverage of Paglen. Laplace-plane frozen-orbit arguments (Rosengren & Scheeres 2014) give theoretical average-stability support but are not gigayear propagations.",
   "vote": "merged [6]+[8], both 3-0"
  },
  {
   "claim": "IADC graveyard guideline history: the original IADC-02-01 (2002) GEO disposal guideline specified only a minimum perigee raise (235 km + 1000*CR*A/m) with no eccentricity clause; Lewis et al. 2004 showed this is insufficient alone and must be paired with an initial-eccentricity constraint. The regulator validated this: IADC-02-01 Rev 1 (Sept 2007) added 'eccentricity less than or equal to 0.003' as condition 2, retained through Rev 2 (2020) and Rev 3 (2021). Frame historically \u2014 the gap was identified and closed in 2007.",
   "confidence": "medium",
   "sources": [
    "https://www.sciencedirect.com/science/article/abs/pii/S0094576504001924",
    "https://www.unoosa.org/pdf/limited/c1/AC105_C1_L260E.pdf"
   ],
   "evidence": "Paper: 'the IADC guideline specifying a preferred initial perigee is appropriate if it accompanies a guideline for the initial eccentricity of the disposal orbit.' Verifier confirmed original IADC-02-01 via UNCOPUOS A/AC.105/C.1/L.260 sec 5.3.1 (perigee-only) and Rev 1 sec 5.3.1 (adds e<=0.003). Medium confidence assigned due to the 2-1 verification vote, though the underlying regulatory documents are primary.",
   "vote": "2-1 [7]"
  },
  {
   "claim": "Chaotic GEO derelicts: uncontrolled geostationary satellites abandoned near either unstable equilibrium point of the GEO ring (165E and 15W) undergo irregular, effectively unpredictable transitions between dynamical states (continuous circulation, long libration, short libration) \u2014 long-term positional forecasting is impractical. Mechanism: interaction of tesseral geopotential harmonics (longitudinal dynamics) with orbital-plane precession forced by Earth's oblateness plus lunisolar perturbations \u2014 the same luni-solar mechanism driving the ~53-yr, 0-15 deg GEO inclination oscillation cycle; chaos onset concentrates at points of the precession cycle when inclination is minimal. Objects librating deep around the stable points (~75E/105W) behave regularly.",
   "confidence": "high",
   "sources": [
    "https://arxiv.org/abs/2510.12227",
    "https://doi.org/10.2514/1.G007467"
   ],
   "evidence": "Flores, Daquin, Pontani, Susanto, Fantino (Nonlinear Dynamics, 2026; arXiv 2510.12227) verbatim: transitions 'extremely sensitive to small perturbations, making the long-term evolution unpredictable'; 'caused by the interaction between the longitudinal dynamics, governed by the tesseral harmonics of the geopotential, and the orbital precession forced by Earth's oblateness and lunisolar perturbations.' Peer-reviewed companion: Flores et al., JGCD 46(10):1919-1928 (2023). Prior corroboration: Celletti et al. (arXiv:1612.08849) on 1:1 tesseral resonance separatrix chaos. Western unstable point varies ~11.5-15W by geopotential model.",
   "vote": "merged [9]+[10], both 3-0"
  },
  {
   "claim": "Lunar lava tubes / pit caves as archive vaults: caves extending from lunar pits would behave as blackbody cavities at ~290 K in radiative equilibrium, nearly isothermal far from the opening, varying by less than 1 C over an entire lunar day (Mare Tranquillitatis pit-cave models) \u2014 versus surface diurnal swings of roughly 100-400 K. Inside these cavities, exposure to cosmic rays, solar radiation, micrometeorites, and thermal extremes is substantially reduced; independent radiological work (J. Radiol. Prot. 2020) finds >1200 g/cm^2 (~7.5 m basalt) roof cover reduces GCR+SEP dose below Earth-surface levels. Caveats: model predictions (COMSOL 2D/3D + 11 yr Diviner data), not in-cave measurements; applies to low-latitude pits and shadowed interiors (sunlit pit floors exceed 420 K at noon); cave existence at MTP later supported by Carrer et al. 2024 (Nature Astronomy) radar.",
   "confidence": "high",
   "sources": [
    "https://luna1.diviner.ucla.edu/~dap/pubs/113.pdf",
    "https://doi.org/10.1029/2022GL099710"
   ],
   "evidence": "Horvath, Hayne & Paige 2022 (GRL, peer-reviewed, Diviner team) verbatim: 'Caves stemming from lunar pits would behave like blackbody cavities at ~290 K'; 'varying by less than 1\u00b0C over an entire lunar day'; 'Inside these cavities exposure to cosmic rays, solar radiation, micrometeorites, and harsh temperature extremes would be substantially reduced.' Analytic check: Teq = (F cos phi / pi sigma)^1/4 \u2248 296 K at equator. 'Archive vault suitability' is a mild extrapolation from the paper's habitation framing, but the protection statement is directly sourced.",
   "vote": "merged [11]+[12]+[13], all 3-0"
  },
  {
   "claim": "DNA storage economics: as of 2022, DNA synthesis (write) cost >$100,000/GB and sequencing (read) >$500/GB; IARPA's MIST program targeted 1 TB/system at $1/GB by 2025 for enterprise archival use (write 1 TB + read 10 TB per day at <$1,000, which reconciles to $1/GB). This is the authoritative anchor for the DNA row of the cost matrix \u2014 a 5-orders-of-magnitude gap between demonstrated and target write cost.",
   "confidence": "high",
   "sources": [
    "https://www.nationalacademies.org/read/27445/chapter/4"
   ],
   "evidence": "NASEM 2024 Rapid Expert Consultation on Archival Data Storage, quoting IARPA's David Markowitz (SC22, 2022): 'DNA synthesis cost at more than $100 k/GB and sequencing cost at more than $500/GB... IARPA MIST 2025 goal: 1 TB/system at $1/GB.' Corroborated by IARPA MIST program materials and GenomeWeb. Correctly framed as dated (2022 costs, 2025 target); whether MIST hit the target is unverified.",
   "vote": "3-0 [14]"
  },
  {
   "claim": "Microsoft Project Silica: raw capacity upward of 7 TB in a 120 x 120 x 2 mm glass platter, estimated TRL 6 (NASEM panel, Jan 2024). Status caveat: Microsoft's Feb 2026 paper reports demonstrated per-tablet capacities of 2.02 TB (Gen-2 borosilicate) and ~4.8-5.15 TB (fused silica) \u2014 consistent with 7 TB being raw pre-ECC capacity \u2014 and Microsoft stated the research phase is complete with no productization. No public cost/GB was found.",
   "confidence": "high",
   "sources": [
    "https://www.nationalacademies.org/read/27445/chapter/4"
   ],
   "evidence": "NASEM report verbatim: 'raw capacity upward of 7 TB in a square glass platter of 120 mm \u00d7 120 mm \u00d7 2 mm... Silica data storage has an estimated TRL of 6.' Corroborated by Tom's Hardware and TechRadar (4 TB and 7 TB densities). Blocks & Files (19 Feb 2026) documents lower demonstrated per-tablet numbers and end of research phase \u2014 flag the 7 TB as raw/estimated and the project status as evolved.",
   "vote": "3-0 [15]"
  },
  {
   "claim": "Cerabyte ceramic-on-glass: 125 GB per medium at 100 nm bit size (~100 PB/rack), roadmap to 1 TB+ per medium at 30 nm (~1 EB/rack), cost projected below $1/TB by end of decade (vs tape ~$2/TB). These are vendor roadmap TARGETS relayed via the NASEM consultation (TRL 6 demo Oct 2023), not demonstrated capability: 2025-26 coverage cites current ~300 nm bits, pilot slipped to 2025-26 at 1 PB/rack, and the company is seed-stage (~$7.7-17M raised; In-Q-Tel, Western Digital, Pure Storage investors) \u2014 real execution risk on the cost target.",
   "confidence": "high",
   "sources": [
    "https://www.nationalacademies.org/read/27445/chapter/4",
    "https://snia.org/sites/default/files/2025-05/SNIA-SDC23-Pflaum-Permanent-Storage_3.pdf"
   ],
   "evidence": "NASEM verbatim: '125 GB per data medium with 100 nm bit size... 1 TB+ per data medium with 30 nm bit size... less than $1 per terabyte by the end of this decade,' rack density 100 PB scaling to 1 EB. Corroborated by Cerabyte's SNIA SDC23 deck and 2025-26 press (Tom's Hardware, TechPowerUp). High confidence in what is claimed/projected; the projections themselves are unproven.",
   "vote": "3-0 [16]"
  },
  {
   "claim": "piqlFilm (vendor specs): 120 GB per reel across ~65,000 film frames (950 m reel, ~2 MB per 35mm frame, 1180 line pairs/mm nano silver halide); Piql claims >1000-year storage life on a PET base with gelatine emulsion containing silver halides, citing unnamed third-party labs and 15 years of accelerated aging \u2014 no peer-reviewed citation on the page and no pricing published. Independent qualifier: Norner's accelerated-aging test reportedly found ~750 years at 21 C/50% RH, longer in cold storage. Consistent with round-1 GitHub Arctic Code Vault figures (21 TB / 186 reels \u2248 113 GB/reel).",
   "confidence": "high",
   "sources": [
    "https://piql.com/technology/film/",
    "https://www.itpro.com/infrastructure/server-storage/358137/deep-freeze-data-that-will-last-1000-years"
   ],
   "evidence": "Live fetch (July 2026) verbatim: 'total storage capacity of 120GB per film,' 'approximately 65,000 frames of film,' 'data can be stored on piqlFilm for more than 1000 years,' PET + silver halide description. Minor internal vendor inconsistency: 120 GB / 2 MB per frame implies ~60,000 frames vs the stated ~65,000. All longevity figures are vendor claims; label as such in the matrix.",
   "vote": "merged [17]+[18], both 3-0"
  },
  {
   "claim": "NanoRosetta / nickel analog media (vendor specs): a 160 mm nickel wafer holds 'up to over 81,000 pages' on a usable annulus of radius 10-60 mm (~737 pages/cm^2 over the annulus, ~403 pages/cm^2 whole-wafer average \u2014 about 3x lower density than the round-1 Nanofiche 2,000 pages/cm^2 claim); typical jobs are far lower ('ten thousand pages or more would not be unusual'). Longevity claimed as 'up to thousands of years' (elsewhere '10,000+ years' and heat 'to 2,647 F' \u2014 merely nickel's melting point, 1,455 C) with resistance to water and EMP, but NO test standard, quantified threshold, or third-party verification cited anywhere on the site. Mastering process: laser-etch glass wafer in polar raster, then electroform a 'father' nickel master (standard optical-disc stamper chain; Bruce Ha patents US 8,387,529 / 8,069,782); NO USD cost per master is published.",
   "confidence": "high",
   "sources": [
    "https://nanorosetta.com/technology/"
   ],
   "evidence": "Live fetch 2026-07-05 verbatim: 'A 160 mm nickel wafer can hold up to over 81,000 pages'; annulus 'from radius 10mm to a maximum of 60mm'; 'Nickel has a much longer life span than microfilm, up to thousands of years... resistant to high temperatures, water damage, and electromagnetic disruption'; 'electroforming creates father nickel master from the inscribed glass wafer.' Arithmetic verified (81,000 / 110 cm^2 = ~737/cm^2). All performance figures are vendor marketing maxima; legibility at max density (~500-700x linear reduction) unstated.",
   "vote": "merged [19]+[20]+[21], all 3-0"
  },
  {
   "claim": "SpaceX rideshare pricing (official, current July 2026): $350,000 base for 50 kg to sun-synchronous orbit, additional mass at $7,000/kg; rates also offered to mid-inclination LEO, GTO, and TLI (quote-based). SpaceX's own pricing API (api-rideshare.spacex.com/api/v2/price) publishes a year-indexed escalator: $5,500/kg (2023), $6,000 (2024), $6,500 (2025), $7,000 (2026), $7,500/kg (2027) \u2014 i.e., ~$500/kg/yr increases, consistent with the 2026 headline. These are advertised base prices; integration, adapter, and licensing costs are extra. (API field 'pricePerKilogramUsd' returns values in cents \u2014 700000 = $7,000 \u2014 verified by internal consistency and the public price card.)",
   "confidence": "high",
   "sources": [
    "https://www.spacex.com/rideshare",
    "https://api-rideshare.spacex.com/api/v2/price"
   ],
   "evidence": "Live API fetch 2026-07-05 (SpaceX's own Fastly CDN): pricingMenus {2023: 550000... 2027: 750000}. Price-card text: 'COST AS LOW AS $350k \u2014 $350k for 50kg to SSO with additional mass at $7k/kg.' Corroborated by New Space Economy (Feb 2026, updated Jun 2026), SatBase (Feb 2026), and Payload's 2023 report of announced annual increases. Direct page render was JS-blocked; exact wording rests on search-index snippet plus two secondary sources within ~1 week of fetch date.",
   "vote": "merged [22]+[23], both 3-0"
  }
 ],
 "caveats": "Coverage gaps dominate: section F (C2PA, RFC 3161/ETSI LTV, RFC 4998, FIPS 204/205, Sigstore/Rekor, Ostrom principles, archive-capture case studies) produced zero surviving claims, and large parts of A/D/E (Voyager/Pioneer/LAGEOS/KEO/MoonArk/Lunar Codex/Peregrine 2024/Sanctuary, Celestis/LifeShip exact pricing, GEO hosted-payload and lunar $/kg costs, Long Now/Internet Archive/Wikimedia/Software Heritage/Svalbard budgets) are unverified this round \u2014 treat those topics as open, not settled. Vendor-claim vs peer-reviewed labeling matters: Piql, NanoRosetta, MOM, and Cerabyte figures are self-reported or roadmap targets (NanoRosetta cites no test standard at all; Cerabyte's <$1/TB is a seed-stage projection; MOM's burst-pressure figure has no published engineering analysis and '2 km deep' means tunnel distance, not overburden); by contrast, the orbit-dynamics and lunar-cave findings are peer-reviewed but are model outputs (200-yr propagations, no in-cave measurements). One claim was refuted in verification: the '~55-year Lyapunov time for Galileo graveyard orbits' figure (1-2 vote) must not be cited, though the broader MEO-instability finding stands. The IADC eccentricity gap should be framed historically (closed by Rev 1 in 2007, e<=0.003). Time-sensitivity: SpaceX prices escalate ~$500/kg annually and were fetched 2026-07-05; Project Silica's status changed materially in Feb 2026 (research phase complete, demonstrated 2-5 TB/tablet vs 7 TB raw); Beresheet Lunar Library survival remains an unfalsified foundation self-assessment. DNA costs are anchored to 2022 data; whether IARPA MIST hit its 2025 $1/GB goal is unknown.",
 "openQuestions": [
  "Section F is entirely unanswered: what is the 2025-2026 adoption status of C2PA, long-term signature validation (ETSI AdES/LTV, RFC 4998 evidence records), post-quantum signatures (FIPS 204/205, SPHINCS+) and transparency-log timestamping for century-scale archival integrity, and which Ostrom-style governance structures (Wikimedia/IA/Software Heritage) best resist capture?",
  "Funding-model verification beyond MOM: exact Celestis and LifeShip pricing tiers, Long Now / Internet Archive / Wikimedia / Software Heritage annual budgets and funding mixes, and the Norway-plus-Crop-Trust structure behind Svalbard \u2014 none survived this round.",
  "Deep-space and lunar delivery economics: verified $/kg for CLPS lunar landers (Astrobotic ~$1.2M/kg?, Intuitive Machines, Firefly), GEO hosted-payload pricing (Astranis/SES/Intelsat), and payload qualification costs (NASA GEVS vibration, ASTM E595 outgassing) remain unquantified \u2014 only LEO rideshare pricing is nailed down.",
  "Preservation-attempts survey remainder: Voyager Golden Record longevity estimates, Pioneer plaques, LAGEOS-1's ~8.4 Myr orbit, KEO's failure history, Sanctuary/Arnano sapphire disk status, MoonArk, Lunar Codex, and the January 2024 Peregrine payload loss all still need verified sourcing; did IARPA MIST reach its 2025 DNA cost target?"
 ],
 "refuted": [
  {
   "claim": "Even an orbit declared 'stable/safe' for Galileo graveyard disposal exhibits a Lyapunov (predictability horizon) time of only about 55 years, and many MEO disposal orbits studied have Lyapunov times of only a few decades \u2014 meaning long-term (century-to-millennium) trajectory predictions in MEO are fundamentally unreliable for archive placement.",
   "vote": "1-2",
   "source": "https://academic.oup.com/mnras/article/464/4/4063/2527864"
  }
 ],
 "sources": [
  {
   "url": "https://www.space.com/predicting-voyager-golden-records-distant-future",
   "quality": "secondary",
   "angle": "Primary survey: space and terrestrial time-capsule projects, outcomes and failure post-mortems",
   "claimCount": 5
  },
  {
   "url": "https://en.wikipedia.org/wiki/KEO",
   "quality": "secondary",
   "angle": "Primary survey: space and terrestrial time-capsule projects, outcomes and failure post-mortems",
   "claimCount": 5
  },
  {
   "url": "https://www.archmission.org/spaceil",
   "quality": "primary",
   "angle": "Primary survey: space and terrestrial time-capsule projects, outcomes and failure post-mortems",
   "claimCount": 5
  },
  {
   "url": "https://en.wikipedia.org/wiki/Peregrine_Mission_One",
   "quality": "secondary",
   "angle": "Primary survey: space and terrestrial time-capsule projects, outcomes and failure post-mortems",
   "claimCount": 5
  },
  {
   "url": "https://en.wikipedia.org/wiki/LAGEOS",
   "quality": "secondary",
   "angle": "Primary survey: space and terrestrial time-capsule projects, outcomes and failure post-mortems",
   "claimCount": 5
  },
  {
   "url": "https://www.memory-of-mankind.com/",
   "quality": "primary",
   "angle": "Primary survey: space and terrestrial time-capsule projects, outcomes and failure post-mortems",
   "claimCount": 5
  },
  {
   "url": "https://academic.oup.com/mnras/article/464/4/4063/2527864",
   "quality": "primary",
   "angle": "Academic/technical: long-term orbital stability and space-environment physics for artifact placement",
   "claimCount": 5
  },
  {
   "url": "https://www.sciencedirect.com/science/article/abs/pii/S0094576504001924",
   "quality": "primary",
   "angle": "Academic/technical: long-term orbital stability and space-environment physics for artifact placement",
   "claimCount": 5
  },
  {
   "url": "https://arxiv.org/abs/2510.12227",
   "quality": "primary",
   "angle": "Academic/technical: long-term orbital stability and space-environment physics for artifact placement",
   "claimCount": 5
  },
  {
   "url": "https://luna1.diviner.ucla.edu/~dap/pubs/113.pdf",
   "quality": "primary",
   "angle": "Academic/technical: long-term orbital stability and space-environment physics for artifact placement",
   "claimCount": 5
  },
  {
   "url": "https://www.nationalacademies.org/read/27445/chapter/4",
   "quality": "primary",
   "angle": "Materials cost matrix: archival media specs and $/GB",
   "claimCount": 5
  },
  {
   "url": "https://www.blocksandfiles.com/2025/12/15/optical-upstarts-chase-century-long-archive-storage/1712734",
   "quality": "secondary",
   "angle": "Materials cost matrix: archival media specs and $/GB",
   "claimCount": 5
  },
  {
   "url": "https://www.datacenterdynamics.com/en/news/ceramic-long-term-storage-company-cerabyte-secures-investment-from-western-digital/",
   "quality": "secondary",
   "angle": "Materials cost matrix: archival media specs and $/GB",
   "claimCount": 5
  },
  {
   "url": "https://www.tomshardware.com/pc-components/storage/project-silica",
   "quality": "secondary",
   "angle": "Materials cost matrix: archival media specs and $/GB",
   "claimCount": 5
  },
  {
   "url": "https://piql.com/technology/film/",
   "quality": "primary",
   "angle": "Materials cost matrix: archival media specs and $/GB",
   "claimCount": 5
  },
  {
   "url": "https://nanorosetta.com/technology/",
   "quality": "primary",
   "angle": "Materials cost matrix: archival media specs and $/GB",
   "claimCount": 5
  },
  {
   "url": "https://www.spacex.com/rideshare",
   "quality": "primary",
   "angle": "Delivery economics: rideshare price lists, CLPS lunar $/kg, hosted payloads, retail keepsake tiers",
   "claimCount": 5
  },
  {
   "url": "https://payloadspace.com/payload-research-the-ultra-low-cost-economics-of-nasas-clps-lunar-program/",
   "quality": "secondary",
   "angle": "Delivery economics: rideshare price lists, CLPS lunar $/kg, hosted payloads, retail keepsake tiers",
   "claimCount": 5
  },
  {
   "url": "https://www.celestis.com/experiences-pricing/",
   "quality": "primary",
   "angle": "Delivery economics: rideshare price lists, CLPS lunar $/kg, hosted payloads, retail keepsake tiers",
   "claimCount": 5
  },
  {
   "url": "https://www.astrobotic.com/lunar-delivery/send-to-the-moon/",
   "quality": "primary",
   "angle": "Delivery economics: rideshare price lists, CLPS lunar $/kg, hosted payloads, retail keepsake tiers",
   "claimCount": 5
  },
  {
   "url": "https://lifeship.com/pages/products",
   "quality": "primary",
   "angle": "Delivery economics: rideshare price lists, CLPS lunar $/kg, hosted payloads, retail keepsake tiers",
   "claimCount": 4
  },
  {
   "url": "https://projects.propublica.org/nonprofits/organizations/943242767",
   "quality": "primary",
   "angle": "Governance and integrity: living-archive budgets, provenance standards, and capture failure modes",
   "claimCount": 5
  },
  {
   "url": "https://wikimediafoundation.org/who-we-are/financial-reports/",
   "quality": "primary",
   "angle": "Governance and integrity: living-archive budgets, provenance standards, and capture failure modes",
   "claimCount": 5
  },
  {
   "url": "https://en.wikipedia.org/wiki/Software_Heritage",
   "quality": "secondary",
   "angle": "Governance and integrity: living-archive budgets, provenance standards, and capture failure modes",
   "claimCount": 5
  },
  {
   "url": "https://www.rfc-editor.org/rfc/rfc4998",
   "quality": "primary",
   "angle": "Governance and integrity: living-archive budgets, provenance standards, and capture failure modes",
   "claimCount": 5
  },
  {
   "url": "https://c2pa.org/wp-content/uploads/sites/33/2025/10/content_credentials_wp_0925.pdf",
   "quality": "primary",
   "angle": "Governance and integrity: living-archive budgets, provenance standards, and capture failure modes",
   "claimCount": 5
  },
  {
   "url": "https://en.wikipedia.org/wiki/Hachette_v._Internet_Archive",
   "quality": "secondary",
   "angle": "Governance and integrity: living-archive budgets, provenance standards, and capture failure modes",
   "claimCount": 5
  }
 ],
 "stats": {
  "angles": 5,
  "sourcesFetched": 27,
  "claimsExtracted": 134,
  "claimsVerified": 25,
  "confirmed": 24,
  "killed": 1,
  "afterSynthesis": 14,
  "urlDupes": 0,
  "budgetDropped": 3,
  "agentCalls": 109
 }
}