--- slug: the-rubber-second passed: 2026-11-08 by: claude-gifted-feynman-h3nw25, one agent on oversight/claims-pass.md at 85881b22 (HEAD moved to 84e45b39 during the run, which it also checked); layer not yet deployed, reader check on local copies; every finding re-read against its source by the author before fixing checked: 122 confirmed: 104 wrong: 7 unverifiable: 2 observed: 9 stale: 0 --- Live page not checked: the layer was not deployed at the time (404 for the page, /checks/verify-the-rubber-second.mjs, data.json, engine.mjs and the card; two deployed strata returned 200 as a control). Reader check on local copies: `AW_DIR= node verify-the-rubber-second.mjs` exit 0, 418 passed (B skipped outside the repository), F 101/0; `--mutate` 18 of 18 caught; `--online` exit 0, ATNF still serving the pinned 2.9.0 package (SHA-256 55daa29f...), B online 112/0; in the repository 418 + B 162 + F 101. The agent recomputed every date in the text and both tables with its own Python from the committed catalogue lines (to the day), the identification search (11 unique; 3, 2, 0), the 11-, 18- and 37-day scans and the TCB control. `node assay/sources.mjs the-rubber-second --online` was run; its NOT FOUND for the PRESTO entries was GitHub refusing the session (a git fetch of commit 5bb0bfac confirmed both), and for the Wikipedia entries it searched rendered HTML for wikitext (all found in action=raw). Step 6 found a real hole (finding 1): figures changed in one sentence passed. After the fixes, at the commit that lands this record: the verifier ties every figure in the prose to its own sentence and sweeps number words as well as digits; all twelve of the agent's one-sentence edits now exit 1; 455 + B 162 + F 101 passed, `--mutate` 22 of 22 caught; the source record grew to 109 entries. ## Claims - CONFIRMED | The plaques mounted on Pioneer 10 and 11 | https://en.wikipedia.org/w/index.php?title=Pioneer_plaque&action=raw | "placed on board the 1972 Pioneer 10 and 1973 Pioneer 11 spacecraft" - CONFIRMED | the covers of the gold-plated records on Voyager 1 and 2, carry the same diagram | https://en.wikipedia.org/w/index.php?title=Voyager_Golden_Record&action=raw | "The drawing in the lower left-hand corner of the cover is the pulsar map previously sent as part of the plaques on Pioneers 10 and 11." - CONFIRMED | fifteen lines from one point, which is the Sun. The longest runs to the centre of the Galaxy. | https://en.wikipedia.org/w/index.php?title=Pioneer_plaque&action=raw | "15 lines emanating from the same origin"; "This line indicates the Sun's relative distance to the center of the galaxy" - CONFIRMED | that pulsar's period, counted in units of the period of the hydrogen atom's radio line | https://en.wikipedia.org/w/index.php?title=Pioneer_plaque&action=raw | "periods of pulsars, using the hydrogen spin-flip transition frequency as the unit" - CONFIRMED | which the plaque draws at its top left | https://en.wikipedia.org/w/index.php?title=Pioneer_plaque&action=raw | "At the top left of the plaque is a schematic representation of the hyperfine transition of hydrogen" - CONFIRMED | The plaque was designed by Carl Sagan and Frank Drake and drawn by Linda Salzman Sagan | https://images-assets.nasa.gov/image/ARC-1972-AC72-1338/metadata.json | "designed by Drs. Carl Sagan and Frank Drake, the artwork was prepared by Linda Salzman Sagan"; Wikipedia agrees - CONFIRMED | the paper the three of them wrote about it gives that line as 1,420,405,752 cycles a second | https://oeis.org/A350136 | "The value of the hyperfine transition period used as multiplier (1420405752) is also given in that paper." - OBSERVED | A long dash is a zero and a short stroke is a one, and the digits run outward from the Sun. | research/the-rubber-second/READINGS.md (recreated from the NASA original) | B1642-03 read outward = 100000110100101010001110101100 = 550675372 = OEIS a(1) - CONFIRMED | NASA Ames Research Center, image ARC-1972-AC72-1338 | https://images-assets.nasa.gov/image/ARC-1972-AC72-1338/metadata.json | "AVAIL:Title": "ARC-1972-AC72-1338", Center ARC - CONFIRMED | NASA's photograph of the finished artwork is dated 31 January 1972 | https://images-assets.nasa.gov/image/ARC-1972-AC72-1338/metadata.json | "AVAIL:DateCreated": "1972-01-31T00:00:00-08:00" - CONFIRMED | at their true directions on the sky (galactic longitude, counted round from the line to the centre) | https://oeis.org/A350136 | "arranged counterclockwise ... in order of increasing galactic longitude, starting at the 3 o'clock position" - CONFIRMED | On the plaque their lengths give distances | https://www.johnstonsarchive.net/astro/pulsarmap.html | "The map identifies the distance and direction of the 14 pulsars" - OBSERVED | Two lines, B1642-03 and B0329+54, were also read directly off NASA's photograph above, digit by digit, and match. | research/the-rubber-second/READINGS.md; recreated with the crops 520x190+990+880 and 270x220+700+790 | both lines read as in plaque.json - CONFIRMED | The fourteen integers are those listed in OEIS A350136, whose comment says they match ... Table 1 and the binary on the plaque | https://oeis.org/A350136 | "the Table 1 constants (in decimal) exactly match those on the plaque (in binary)" - CONFIRMED | The binary strings are W. R. Johnston's transcription | https://www.johnstonsarchive.net/astro/pulsarmap.html | Table 1; all 14 strings and integers equal data.json - CONFIRMED | "With 14 periods, almost all of which are accurate to nine significant figures in decimal equivalent, ... to the equivalent of the year 1971." | https://www.johnstonsarchive.net/astro/pulsarmap.html | verbatim in Johnston's quotation of the paper - CONFIRMED | Carl Sagan, Linda Salzman Sagan and Frank Drake, A Message from Earth, Science 175, 1972 | https://api.crossref.org/works/10.1126/science.175.4024.881 | Science 175(4024) 881-884, 25 Feb 1972, the same three authors - OBSERVED | We could not read the designers' paper, and nothing quoted from it says where their periods came from. | research/the-rubber-second/sources/NOTES.md (the access log) | True of the author's search (paywalled; Crossref has no reference list). The agent found a copy posted on a university course page, which the layer does not use (finding 9); the page now says "We have not read". - UNVERIFIABLE | W. R. Johnston did this in 2003 ... and got 1969.7, give or take 1.2 years, from the other thirteen. | Johnston's page as updated 30 October 2007; Wayback capture of 2003-12-28 refused to curl and WebFetch | The 2007 page says "this gives a date of origin of 1969.7 ± 1.2 year!" (Table 5 recomputes to 1969.75 ± 1.22) and notes a 2007 correction to pulsar #1; the 2003 text could not be read. - CONFIRMED | W. R. Johnston did this in 2003 with the catalogue of the day | https://www.johnstonsarchive.net/astro/pulsarmap.html | "Using the ATNF Pulsar Catalog (2002), which lists 1,480 pulsars"; "© 2003, 2007" - UNVERIFIABLE | Johnston printed the differences in 2003; he did not explain them. | Wikipedia's reference "updated 11 March 2003" and Johnston's 2007 page | The reference shows only that the page existed in 2003; the 2007 text has Table 2 and only "note that most are quite good." - CONFIRMED | He also set the plaque's periods beside a 1975 catalogue and printed the differences, eight of them between 0.024 and 0.032 parts per million | https://www.johnstonsarchive.net/astro/pulsarmap.html | Table 2: 0.025, 0.024, 0.030, 0.032, 0.030, 0.031, 0.030, 0.030 (the other six 19.38, 1233.0, 105.9, 0.384, 969.2, 0.121) - CONFIRMED | with the remark that "most are quite good" | https://www.johnstonsarchive.net/astro/pulsarmap.html | "note that most are quite good." - CONFIRMED | W. R. Johnston, Reading the Pioneer/Voyager Pulsar Map, updated 30 October 2007. | https://www.johnstonsarchive.net/astro/pulsarmap.html | "updated 30 October 2007" - CONFIRMED | Johnston found the directions somewhat helpful and the distances not. | https://www.johnstonsarchive.net/astro/pulsarmap.html | "The direction information is somewhat helpful, but the distance information is not." - CONFIRMED | Sagan's table, as Johnston transcribed it, gives 0.5579533900 | https://www.johnstonsarchive.net/astro/pulsarmap.html | Table 6: "5.579533900x10-1" - CONFIRMED | The ATNF Pulsar Catalogue 2.9.0 ... holds 4,503 pulsars | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | SHA-256 55daa29f..., fetched again by the agent; 4503 PSRJ records in psrcat.db - CONFIRMED | packaged in October 2026 | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | psrcat.db dated 2026-10-07 00:27 in the tar listing - CONFIRMED | The ATNF Pulsar Catalogue 2.9.0 | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | "#CATALOGUE 2.9.0" - CONFIRMED | "The new pulsar was discovered on August 26, 1968", at Arecibo, in search data taken some time before | https://www.nature.com/articles/2191237a0 | verbatim; "Arecibo Ionospheric Observatory"; "search data taken some time earlier" - CONFIRMED | H. D. Craft, J. M. Sutton, J. M. Comella and R. V. E. Lovelace, New Pulsar in Vulpecula, Nature 219 (1968) 1237 | https://api.crossref.org/works/10.1038/2191237a0 | four authors; Nature 219(5160) 1237-1238, September 1968 - CONFIRMED | The first one's signal was recorded in August 1967 | https://en.wikipedia.org/w/index.php?title=Pulsar&action=raw | "data recorded on August 6, 1967" - CONFIRMED | recognised as pulses that November | https://en.wikipedia.org/w/index.php?title=Pulsar&action=raw | "On November 28, 1967, ... resolved the signals as a series of pulses" - CONFIRMED | a spinning neutron star whose radio beam sweeps past us once a turn | https://en.wikipedia.org/w/index.php?title=Pulsar&action=raw | "observed only when a beam of emission is pointing toward Earth (similar to ... a lighthouse)" - CONFIRMED | the SI second the world has kept since 1967 | https://tf.nist.gov/general/pdf/1796.pdf (printed p. 110) | "In 1967, the second was defined in terms of the frequency of radiation emitted by a cesium atom" - CONFIRMED | From 1961 to 1971 the world's broadcast time ... did not tick in atomic seconds. | https://maia.usno.navy.mil/ser7/tai-utc.dat | "1961 JAN 1 ... X 0.001296 S"; NASA TN D-7065 p. 2 agrees - CONFIRMED | USNO's table ... from 1 January 1966 to the end of 1971 ... grew by 0.002592 seconds every day | https://maia.usno.navy.mil/ser7/tai-utc.dat | "1966 JAN 1 ... X 0.002592 S" ... "1972 JAN 1 ... X 0.0 S" - CONFIRMED | So its second was set a little longer than the atomic one ... to keep pace with the Earth's turning. | https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nbsspecialpublication236e1970.pdf | p. 3 (c): "intentionally offset ... to reduce departure between the time signals as broadcast and astronomical time, UT2" - CONFIRMED | by an amount chosen in advance | https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nbsspecialpublication236e1970.pdf | p. 3 (c): "a particular offset ... will remain in effect for the entire calendar year" - CONFIRMED | "International Morse Code symbols for M300, ... voice announcement." | https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nbsspecialpublication236e1970.pdf | p. 3 (d), verbatim on the page image - CONFIRMED | The 300 is parts in ten thousand million: the same 3 parts in a hundred million. | https://nvlpubs.nist.gov/nistpubs/Legacy/SP/nbsspecialpublication236e1970.pdf | "in 1966 through 1970, -300 parts in 10^10" - WRONG | "But after a few years ... at a different rate." (footer: NIST Monograph 155, 1999, p. 110) | https://tf.nist.gov/general/pdf/1796.pdf | The quotation is verbatim but on printed p. 109 (PDF p. 112). - CONFIRMED | NIST's history of timekeeping calls the arrangement the rubber second | https://tf.nist.gov/general/pdf/1796.pdf | heading "'Rubber' Seconds" (p. 108); "the 'rubber' second came into being" (p. 109) - CONFIRMED | It ended on 1 January 1972; "eliminates the frequency offset of 300 parts in 10^10 ... (the UTC second) constant" | https://ntrs.nasa.gov/api/citations/19730006465/downloads/19730006465.pdf | abstract verbatim; "implemented ... on January 1, 1972" - CONFIRMED | from then on UTC has been kept near the Earth by whole leap seconds instead | https://ntrs.nasa.gov/api/citations/19730006465/downloads/19730006465.pdf | "by step-time adjustments of exactly 1 s, when needed" - CONFIRMED | A. R. Chi and H. S. Fosque, A Step in Time ..., NASA TN D-7065 (1973) | https://ntrs.nasa.gov/api/citations/19730006465/downloads/19730006465.pdf | documentation page: Chi and Fosque, January 1973 - CONFIRMED | Pioneer 10 left on 3 March 1972. | https://nssdc.gsfc.nasa.gov/nmc/spacecraft/display.action?id=1972-012A | "Launch Date: 1972-03-03" - CONFIRMED | JPL ... "taking into account that P is in UTC seconds" | https://articles.adsabs.harvard.edu/pdf/1970ApJ...159L..35R | p. L38, verbatim - CONFIRMED | The Jet Propulsion Laboratory, timing eleven pulsars | https://articles.adsabs.harvard.edu/pdf/1970ApJ...159L..35R | "Time-of-Arrival Observations of Eleven Pulsars", Jet Propulsion Laboratory - CONFIRMED | Princeton's optical timing of the Crab used "the Universal Time second" | https://articles.adsabs.harvard.edu/pdf/1972ApJ...175..217B | p. 225: "The unit of time used in table 1 is the Universal Time second" - CONFIRMED | Princeton's optical timing of the Crab | https://articles.adsabs.harvard.edu/pdf/1972ApJ...175..217B | Joseph Henry Laboratories, Princeton University - CONFIRMED | Jodrell Bank quoted its periods "in A1 seconds", atomic ones | https://articles.adsabs.harvard.edu/pdf/1971MNRAS.153..119H | p. 125: "P0 is quoted in A1 seconds as defined in Section 6.1" - CONFIRMED | recovered from a copy committed to Scott Ransom's PRESTO software that November | https://github.com/scottransom/presto (commit 5bb0bfac, by git fetch) | 2002-11-18 05:31:11 +0000; lib/psr_export.dat, 1484 lines, byte-identical to the kept copy (fb5ae0d5...) - CONFIRMED | Scott Ransom's PRESTO software | https://github.com/scottransom/presto | author "Scott Ransom " - WRONG | That catalogue's solutions for three of the five go back to timing that began in 1970 itself | research/the-rubber-second/data/atnf-2002.json | Only B0329+54 (40621.0) and B1642-03 (40621.54) of the five have 1970 epochs; B0950+08 is 41500.5, B1929+10 48381, B2217+47 48382. The third 1970 solution is B2016+28, not one of the five. - CONFIRMED | In April 1972, a month after Pioneer 10 left, R. N. Manchester and W. L. Peters ... published | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | "173:221-226, 1972 April 1" - CONFIRMED | published periods for 19 pulsars | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | "periods, period derivatives, and positions for 19 pulsars" - CONFIRMED | of the National Radio Astronomy Observatory | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | "National Radio Astronomy Observatory, Green Bank, West Virginia" - CONFIRMED | they had timed at Green Bank in 1970 and 1971 | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | "in several sessions during 1970 and 1971" - CONFIRMED | received in August 1971 and revised that November | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | "Received 1971 August 23; revised 1971 November 3" - CONFIRMED | Its table heads the periods "P(A.1)" | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | p. 224 image: column head "P(A.1) (s)" - CONFIRMED | their clock, they write, "was adjusted to UTC each session" | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | p. 223: "This clock was adjusted to UTC each session to an accuracy of better than ±10 μs using Loran C" - CONFIRMED | seconds of A.1, the US Naval Observatory's atomic time | https://www.cl.cam.ac.uk/~mgk25/time/metrologia-leapsecond.pdf | section 5.1: "USNO inaugurated its A.1 atomic time scale on 13 September 1956" - CONFIRMED | R. N. Manchester and W. L. Peters, Pulsar Parameters from Timing Observations, ApJ 173 (1972) 221 to 226 | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | title and authors as cited - CONFIRMED | Shorten each ... within a nanosecond; six agree in every printed digit; B1642-03 if cut off; 7 to 21 nanoseconds away | https://articles.adsabs.harvard.edu/pdf/1972ApJ...173..221M | Redone by hand from Table 2 and Johnston's transcription: 0.7145186424, 0.2530650432, 0.3876887797 (779 truncated), 0.226517038, 0.35873542, 0.55795339, 0.538467378; unshortened gaps 7.1 to 21.5 ns - CONFIRMED | Hunt's periods for six of the same pulsars ... agree ... to within a quarter of a nanosecond | research/the-rubber-second/sources/derived/hunt_vs_mp.out.txt | maximum 0.22 ns (B1933+16); Hunt Table I read off the scan - OBSERVED | How the numbers reached the plaque in UTC's seconds ... we could not find out. | research/the-rubber-second/sources/NOTES.md | not kept as a separate record; the searches are listed in the notes - CONFIRMED | Hunt's own Jodrell Bank periods, 0.530659599042 and 3.745490813 seconds | https://articles.adsabs.harvard.edu/pdf/1971MNRAS.153..119H | p. 125 image: "0·530 659 599 042", "3·745 490 813" - CONFIRMED | "quoted in A1 seconds", cut to as many figures as Sagan's table gives and not shortened | https://articles.adsabs.harvard.edu/pdf/1971MNRAS.153..119H | footnote; 9 and 8 figures match unshortened, not shortened - CONFIRMED | the carrier of the MSF transmitter at Rugby, "based on the atomic scale" | https://articles.adsabs.harvard.edu/pdf/1971MNRAS.153..119H | section 3, p. 120: "MSF Rugby, in which the carrier frequency is based on the atomic scale A1" - CONFIRMED | Jodrell Bank's clock ran at the atomic rate | https://articles.adsabs.harvard.edu/pdf/1971MNRAS.153..119H | "The clock rate was approximately equal to A1" - CONFIRMED | Hunt's first observation of it on 21 Jan 1969 | https://articles.adsabs.harvard.edu/pdf/1971MNRAS.153..119H | Table I "Julian Date of first observation" 2440 242·5 = 1969 Jan 21 - CONFIRMED | G. C. Hunt, The Rate of Change of Period of the Pulsars, MNRAS 153 (1971) 119 to 131 | https://articles.adsabs.harvard.edu/pdf/1971MNRAS.153..119H | p. 119: "Mon. Not. R. astr. Soc. (1971) 153, 119-131. THE RATE OF CHANGE OF PERIOD OF THE PULSARS" - CONFIRMED | Princeton's fits ... from its timing of March 1969 to April 1971 | https://articles.adsabs.harvard.edu/pdf/1972ApJ...175..217B | Table 1 span "1969 Mar. 9 - 1971 Apr. 23" - CONFIRMED | from its timing of March 1969 to April 1971 (the data entry) | research/the-rubber-second/data/boynton-1972.json | "1969 Mar. 9-1971 Apr. 23" - CONFIRMED | It is a southern pulsar, found at Molonglo in Australia in 1968. | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | B1451-68's PSRB reference lvw68 = Large, Vaughan and Wielebinski 1968, Nature 220, 753 - CONFIRMED | found at Molonglo in Australia in 1968 (entry b1451-oct68) | https://articles.adsabs.harvard.edu/pdf/1969ApL.....3..123L | The excerpt (on clustering) did not state 1968; the 1968 date rests on lvw68 and on p. 125's "dates back to October 1968" (finding 10) - CONFIRMED | Molonglo's "0.264±.002" seconds, published in 1969 | https://articles.adsabs.harvard.edu/pdf/1969ApL.....3..123L | Table I, p. 125 image: "PSR 1451-68 ... 0.264±.002" - OBSERVED | The only period for it from before the plaque that we could read | research/the-rubber-second/sources/NOTES.md | not kept as a separate record - CONFIRMED | the catalogue's glitch table lists 21 for Vela and 8 for B1727-47 | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | glitch.db: 21 J0835-4510 lines, 8 J1731-4744 lines - OBSERVED | We found no published source for either figure. | research/the-rubber-second/sources/NOTES.md | not kept as a separate record - CONFIRMED | Large, Vaughan and Wielebinski published from Molonglo in 1969, "0.388±.004" | https://articles.adsabs.harvard.edu/pdf/1969ApL.....3..123L | Table I: "MP 1240 ... 0.388±.004" - CONFIRMED | M. I. Large, A. E. Vaughan and R. Wielebinski, ... Astrophysical Letters 3 (1969) 123 | https://articles.adsabs.harvard.edu/pdf/1969ApL.....3..123L | "Astrophysical Letters, 1969, Vol. 3, pp. 123-128" - CONFIRMED | B1240-64 the plaque gives as 0.388 seconds, written out to ten figures with seven zeros | research/the-rubber-second/data/plaque.json | Johnston's transcription "3.880000000e-1"; OEIS a(12) / 1420405752 = 0.3880000002 - CONFIRMED | B1240-64 the plaque gives as 0.388 seconds | https://en.wikipedia.org/w/index.php?title=Pioneer_plaque&action=raw | "known to only three significant decimal digits: 0.388 second" - CONFIRMED | anomalous X-ray pulsars and soft gamma-ray repeaters | https://www.atnf.csiro.au/research/pulsar/psrcat/psrcat_help.html | Appendix B: "Anomalous X-ray Pulsar or Soft Gamma-ray Repeater with detected pulsations" - CONFIRMED | B0525+21 matches 3 (the other two are of the type the catalogue marks AXP | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | J1714-3810 "TYPE AXP,NRAD"; the agent's own search finds 3 - CONFIRMED | which slow so fast that their periods sweep through its value | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | J1745-2900 TYPE AXP, F1 -1.24e-12; J1714-3810 P1 5.88e-11 - CONFIRMED | Searched by period alone, across all 4,503 ... the plaque's numbers pick out 11 of the fourteen uniquely. | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | the agent's own parser: 11 unique; B0525+21 3; B0823+26 2; B1240-64 0 - CONFIRMED | launched in August ... 1977 | https://nssdc.gsfc.nasa.gov/nmc/spacecraft/display.action?id=1977-076A | "Launch Date: 1977-08-20" - CONFIRMED | and September 1977 | https://nssdc.gsfc.nasa.gov/nmc/spacecraft/display.action?id=1977-084A | "Launch Date: 1977-09-05" - CONFIRMED | carry the same map with the same numbers | https://en.wikipedia.org/w/index.php?title=Pioneer_plaque&action=raw | "shared in common with the Voyager Golden Record"; Johnston's binary from the Voyager diagram equals the Pioneer integers, all 14 - CONFIRMED | after both Pioneers had left | https://nssdc.gsfc.nasa.gov/nmc/spacecraft/display.action?id=1973-019A | "Launch Date: 1973-04-06" - CONFIRMED | TCB ... whose second differs from the one Earth clocks keep by 1.55 parts in a hundred million | https://en.wikipedia.org/w/index.php?title=Barycentric_Dynamical_Time&action=raw | "L_B = 1.550519768E-8" - OBSERVED | We found no earlier account that ties the plaque's periods to UTC's stretched second | research/the-rubber-second/sources/prior-art/ | the agent's grep of the kept copies and two web searches (which did return Johnston and OEIS, as positive controls) found none - OBSERVED | the code of three projects that redraw the map | research/the-rubber-second/sources/prior-art/github/ | Codeptor/astrolabe, tgoorden/pulsar-map, tlindsay/voyager kept - CONFIRMED | In 2022 the CGPM decided to let UTC drift further from the Earth's turning | https://www.bipm.org/en/cgpm-2022/resolution-4 | "decides that the maximum value for the difference (UT1-UTC) will be increased in, or before, 2035" - CONFIRMED | "continuous UTC will become effective on 20 May 2027" | https://www.bipm.org/en/cgpm-2026/resolution-5 | verbatim - CONFIRMED | the gap between UTC and the Earth's turning may grow to an hour | https://www.bipm.org/en/cgpm-2026/resolution-5 | "the maximum value for the difference UT1-UTC will be 3 600 seconds (1 hour)" - CONFIRMED | the meeting sat in Versailles from 13 to 15 October 2026 | https://www.bipm.org/en/cgpm-2026 | "from Tuesday 13 to Thursday 15 October 2026 at the Palais des Congrès de Versailles" - WRONG | R. E. Reichley, G. S. Downs and G. A. Morris, Time-of-Arrival Observations of Eleven Pulsars, ApJ 159 (1970) L35 | https://api.crossref.org/works/10.1086/180473 | "P. E." Reichley, as the scan's p. L35 also reads - WRONG | J. Jespersen and J. Fitz-Randolph, ... NIST Monograph 155 (1999), pp. 109 to 111 | https://tf.nist.gov/general/pdf/1796.pdf | The cited material is on printed pp. 108-110. - CONFIRMED | P. E. Boynton, E. J. Groth, D. P. Hutchinson, G. P. Nanos Jr., R. B. Partridge and D. T. Wilkinson, Optical Timing of the Crab Pulsar, NP 0532, ApJ 175 (1972) 217 to 241 | https://articles.adsabs.harvard.edu/pdf/1972ApJ...175..217B | "The Astrophysical Journal, 175:217-241, 1972 July 1"; the same six authors - CONFIRMED | R. A. Nelson and others, The leap second: its history and possible future, Metrologia 38 (2001) 509 to 529 | https://api.crossref.org/works/10.1088/0026-1394/38/6/6 | Metrologia 38(6) 509-529, December 2001 - CONFIRMED | R. N. Manchester, G. B. Hobbs, A. Teoh and M. Hobbs, The ATNF Pulsar Catalogue, AJ 129 (2005) 1993 | https://api.crossref.org/works/10.1086/428488 | AJ 129(4) 1993-2006; the full title is "The Australia Telescope National Facility Pulsar Catalogue" (now cited in full) - CONFIRMED | OEIS A350136, Pulsar periods from the Pioneer/Voyager plaques (1972), R. B. Fowler, 2021 | https://oeis.org/A350136 | "%A _Robert B Fowler_, Dec 15 2021" - CONFIRMED | B2016+28, which the plaque draws at 68 degrees from the line to the centre | https://www.johnstonsarchive.net/astro/pulsarmap.html | Table 1: psr #11 (B2016+28) direction 68 - WRONG | Cropped; nothing else changed. | research/the-rubber-second/READINGS.md (recreated: convert of NASA's original, crop 1680x1680+215+150, resize 1100x1100, compare -metric RMSE) | The image is the crop reduced to 1100x1100 (RMSE 0.9%): cropped and reduced to 65%, colours unchanged. - CONFIRMED | B0525+21 ... glitches blur its record | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | glitch.db: 3 glitches for J0528+2200 - CONFIRMED | the quiet pulsars (the ten ... with no glitch on record bigger than a few parts in a billion) | https://www.atnf.csiro.au/research/pulsar/psrcat/downloads/psrcat_pkg.tar.gz | glitch.db: the largest among the ten is 2.64e-9 (B1642-03) - WRONG | Over all pairs of quiet pulsars, ... a median of 2.87 parts in a hundred million | public/strata/the-rubber-second/data.json (recomputed) | It is over 42 of the 45 pairs; three with near-equal sensitivity are dropped. All 45 give 2.89. - CONFIRMED | for windows of 11 and 37 days the peak is at the same place | research/the-rubber-second/results.json (recomputed) | 11 d: peak 4 at 2.9 and 3.1 (3 at 3.0); 37 d: peak 6 at 2.9; inside the 2.9 to 3.1 band - WRONG | [the check] ... checks every figure in the text against it | verify-the-rubber-second.mjs (recreated: step-6 edits on a scratch copy, AW_DIR= node verify-the-rubber-second.mjs) | Figures written as words, and twelve whitelisted bare years, passed with exit 0 when changed in one sentence. - CONFIRMED | Sagan's table, as Johnston transcribed it (all fourteen periods) | https://oeis.org/A350136 | OEIS: "the Table 1 constants (in decimal) exactly match those on the plaque (in binary)"; each of Johnston's fourteen periods times 1,420,405,752 rounds to the plaque's integer (verifier section A) - CONFIRMED | a period of exactly 0.7 atomic seconds measures 0.699999979 UTC seconds | research/the-rubber-second/analyze.mjs (arithmetic) | 0.7 x (1 - 3e-8) = 0.699999979 - CONFIRMED | a month after Pioneer 10 left / abolished nine weeks before the first plaque left Earth / two years before Pioneer 10 flew and seven before the Voyagers | https://nssdc.gsfc.nasa.gov/nmc/spacecraft/display.action?id=1972-012A | 29 days; 62 days; 1970 to 1972 and 1977 - CONFIRMED | every date in the text and both tables (SI and UTC, today's and the 2002 catalogue; Crab 3 Oct 1971; B1240-64 1409 BC, 0.38848 s in 1970) | research/the-rubber-second/data/atnf-2.9.0.json and data/atnf-2002.json (recomputed in Python) | all reproduced to the day (B1240-64 to the year); Boynton's fit about 6 Oct 1971 by hand - CONFIRMED | left in TCB, B0950+08 falls out of the five | research/the-rubber-second/data/atnf-2.9.0.json (recomputed) | B0950+08 moves to 8 Jul 1969; the largest group becomes 2 - CONFIRMED | Three of them agree on a fortnight in November 1969 | research/the-rubber-second/data/atnf-2.9.0.json (recomputed) | SI dates 16, 16 and 30 Nov 1969 - CONFIRMED | 22 Jan to 5 Feb 1970: 15 days, where in SI seconds they spread over 330; 2002 catalogue 31 Jan to 5 Feb, SI across most of a year | research/the-rubber-second/data/atnf-2002.json (recomputed) | 14.75 d; 329.6 d; 2002 SI 17 Jan to 29 Nov 1969 - OBSERVED | The sets were fixed in the analysis script before the coincidence and the null test were computed; the first look at the dates came before any of that | research/the-rubber-second/README.md | not kept beyond the README: one commit, no earlier history of analyze.mjs - CONFIRMED | In 20,000 such plaques, ... in 39, and within 0.2 ... of UTC's second in none | verify-the-rubber-second.mjs section D (re-runs every trial with its own code) | passes in the agent's runs ## Findings - [fixed] HIGH (step 6): figures changed in one sentence passed the check (bare years were vouched for by any other copy of the year; number words were never swept), while the page said the check "checks every figure in the text against it". The verifier's TYPED table is now one fragment of its own sentence per figure, each required to appear exactly as often as it says and, where the data decides the figure, to hold (about 120 fragments); the sweep now fails on any digit or number word left; four mutants added (a year, a heading's number word, a sentence's number word, a unit), 22 of 22 caught; all twelve of the agent's one-sentence edits now exit 1. The page now says the check "checks every computed figure in the text against it, ties every other figure and number word in the text to its sentence and to where it comes from". - [fixed] MED: "That catalogue's solutions for three of the five go back to timing that began in 1970 itself" was two of the five (re-read in data/atnf-2002.json: B0329+54 at MJD 40621, B1642-03 at 40621.54; B0950+08 at 41500.5). The page now says the catalogue "refers its solutions for two of the five, B0329+54 and B1642-03, to days in February 1970 itself"; the verifier checks which of the five have 1970 epochs. - [fixed] MINOR: the NIST quotation is on printed p. 109, not p. 110 (re-read off the kept pages, whose folios are 108, 109 and 110). The footer now says p. 109. - [fixed] MINOR: the sources list said pp. 109 to 111; it now says pp. 108 to 110. The kept extract was renamed to ...-pp108-110-..., and sources/NOTES.md carries a dated correction where the scout's page numbers were one too high. - [fixed] MINOR: "R. E. Reichley" is P. E. Reichley (scan p. L35; Crossref 10.1086/180473, now kept). Corrected in the sources list; a citation entry added to the source record. - [fixed] MINOR: "Over all pairs of quiet pulsars" was over 42 of the 45 (the rule in analyze.mjs drops pairs whose sensitivities differ by under 1e6 years per unit). The page now says "Over the 42 pairs of quiet pulsars that the second moves apart (the other three pairs move almost in step)", both numbers computed by fill-page.mjs and checked by the verifier. - [fixed] MINOR: the figure was cropped and reduced in size, not only cropped. The caption now says "Cropped and reduced in size; nothing else changed."; READINGS.md records the command, recreated at 0.97% RMSE. - [fixed] MINOR (the two UNVERIFIABLE lines): the 2003 text of Johnston's page could not be read, and his 2007 page says it was corrected that year. The page now says he did this "on a page first posted in 2003 and corrected in 2007, with the catalogue of 2002", and "Johnston printed the differences; he did not explain them." without a year. - [declined] MINOR: the agent found the designers' 1972 paper readable as a JSTOR copy posted on a university course page and suggested firming up the page's hedges from it. The layer does not use copies that are not authorised, so the paper stays unread here; nothing on the page rests on it, and the agent reports that nothing on the page is contradicted by it. The page's wording is now "We have not read the designers' paper, which is behind Science's paywall", which is true of its author, and "which we have not read". - [fixed] MINOR (record hygiene): b1451-oct68 now quotes "dates back to October 1968"; hunt-cite quotes the journal line; boy-span quotes Table 1's span; johnston-2003b was rewritten with the page; the Wikipedia entries point at action=raw, so the online check can find their wikitext; the Nature DOI moved from the quotation entry to the citation entry. Nine entries were added for claims the pass checked that the record had left out (the pairs, the crop, the Reichley, ATNF and NIST citations, Johnston's dates). The record has 109 entries. - [declined] `assay/sources.mjs --online` reported NOT FOUND for the two PRESTO entries because GitHub refused this session's requests; the agent confirmed the commit, its date and author by git fetch, and the kept file is byte-identical. No change needed.