Timeline

2026

1. 2026-11-10Milestone

   Listed on the site after the reader panel: the mathematician and the field expert raised no accuracy objection for four rounds running.

2. 2026-11-07Correction

   The opening said erdosproblems.com still marks the problem open. That site stopped showing whether problems are open or solved in October, so the opening now says that no proof has been confirmed and that the community database of Erdős problems on GitHub still records the problem as open. The page also notes that the site paused new comments and proof claims.

3. 2026-11-07Milestone

   The page said no proof had been confirmed. In October a Lean version of the all-forests proof claim was accepted by both proof checkers of the Palomar registry, so the page now says that, and that neither paper has been refereed and the GitHub database still records the problem as open.

4. 2026-11-07Milestone

   Presented as a poster at the workshop in San Francisco.

5. 2026-11-07Correction

   The status line said the problem is still listed as open on erdosproblems.com. That site stopped showing whether problems are open or solved in October, so the line now says that no proof has been confirmed and that the community database of Erdős problems on GitHub still records the problem as open. The page also notes that the site paused new comments and proof claims.

6. 2026-11-07Milestone

   The status line said no proof had been confirmed. In October a Lean version of the all-forests proof claim was accepted by both proof checkers of the Palomar registry, so the line now says that, and that neither paper has been refereed and the GitHub database still records the problem as open.

7. 2026-11-06Milestone

   "Verify Before You Accept" posted on Zenodo, free to read

   The final version of the paper is open access with a permanent DOI, 10.5281/zenodo.23652450, together with the de-identified tables and the scripts that built its counts.

8. 2026-11-06Correction

   The page said my check held for every tree up to the new limit. My own runs covered the three new sizes and some smaller ones, and the rest of the smaller sizes rest on earlier checks; the page now says I extended the check from the old limit to the new one, and how many trees that added. It also gives the dates of my forum posts and of the two later proof claims, neither of which cites my check.

9. 2026-11-06Correction

   The opening now says vertices, with the dots explained once, and states the test the trees pass. The page now says which sizes my own runs covered, and that the smaller sizes rest on the earlier checks, so it no longer reads as if my runs alone covered every tree up to the new limit. It adds how many trees the three new sizes hold, and the dates of the two later proof claims. A sentence about the false alarm had lost its first words and is whole again.

10. 2026-11-06Correction

    The check box said the tables' maintainer reran the largest game and two others; the page body already said four games, and the box now says so too, counting the later rerun.

11. 2026-11-05Milestone

    Added a row under the title linking to the GitHub commit with the final run and to my pull request to the Erdős problems database, with the dates it was opened and merged in Pacific time.

12. 2026-11-05Milestone

    Added a row under the title linking to the credit on Achim Flammenkamp's octal games page, which dates my work on his tables to July.

13. 2026-11-05Milestone

    Added a row under the title linking to the OEIS history, which shows when the entry was submitted and when it was approved.

14. 2026-11-05Milestone

    Added a row under the title linking to the credit on Achim Flammenkamp's octal games page, which dates my work on his tables to July.

15. 2026-11-02Accepted

    A401463 through A401466 published in the OEIS

    They count shapes built from two more space-filling cells, the 17-faced cell of the Laves graph and one of the 38-faced cells found by Peter Engel in 1981. The number of OEIS entries crediting this work moved from 46 to 50, 26 of them authored.

16. 2026-10-29Correction

    The page credited the earlier record to a paper from April; it was set earlier, in January and March, and the paper only cites it. It said the change Terence Tao merged cites this check and that the database listing links to it; the merged change adds the OEIS tree-count link, and only the pull request's description cites the check. The status line now says one of the two September proof claims is partial.

17. 2026-10-29Correction

    The page said each extra dot means a bit under three times as many trees; at these sizes it is about two and three-quarter times. The opening said the conjecture is about trees and forests without saying that only trees were checked; it now says so. It now names who posted the two September proof claims, says the reruns of the earlier sizes matched the tree counts and found no failures, says only the two largest sizes were split into parts, and dates the forum post.

18. 2026-10-29Correction

    The page said Schmitt's survey covered fewer symmetry types than it did, and that it printed one table per type. It now gives the right number of types and says that each pair of mirror-image types shares a table, so the survey has fewer tables than types.

19. 2026-10-29Correction

    The page said the deposit's README describes the verified set as the deposit's result. The README calls the deposit a documented attempt that moves no bound and notes that the verified set was already public in the OEIS; the page now says that.

20. 2026-10-29Correction

    The post said the old word was replaced in the archive. It was replaced in the archive's title and README and in the OEIS link; the archive's description still uses it, as its file names do, and the post now says so.

21. 2026-10-29Correction

    The notes for specialists said tree-search detectors are now anchored to a marker that normal output cannot contain; the rule in force, as the plain version says, is a match only at the start of a line. The plain version now also says that the flaw behind the August polyhex error had been found and written down in July, which is why the page gives both months.

22. 2026-10-29Milestone

    Defined polyform, cell and the fixed, one-sided and free counts in plain words, explained why twice the free count minus the one-sided count gives the mirror-symmetric shapes, and said near the top which sizes a second program confirms.

23. 2026-10-29Correction

    The page credited the earlier record to a paper from April; it was set earlier, by the forum user JakeMallen on the problem's forum in January and by Brett Reynolds's code and preprint in March, and the paper only cites it. The body still said the merged database change cites this check; it adds the OEIS tree-count link, and the pull request's description cites the check. The page said nobody has proved the conjecture; it now notes the two proof claims posted in September and that the problem is still listed as open.

24. 2026-10-29Correction

    The page said each extra dot means a bit under three times as many trees; at these sizes it is about two and three-quarter times, which is what the table shows. It now also names who posted the two September proof claims.

25. 2026-10-29Milestone

    The page now says who did what (what I asked for and did, what the Claude agents did), says that the separate rescan covered only the two largest sizes of the earlier check, gives the forum comments and the end of the run in date order, and explains the ▲ mark where it first appears. The opening no longer tells the later proof claims; a short dated status line points to them.

26. 2026-10-29Correction

    The page undercounted the OEIS entries for these games by one: every game has an entry, one new and the rest older entries I extended. It also listed a missed finding among the errors as if the finding were wrong; it now says the finding was correct and had gone unnoticed. It now names the fourth game the tables' maintainer reran, and explains the failed database search fully: besides the extra starting term, several games are filed under a related game whose list is shifted.

27. 2026-10-29Correction

    The page gave the expert's reference page as last updated in early September and his furthest run on the largest game as his current point; his page was updated again in late October and goes much further, and the page now says so. It also said the final results were recomputed from scratch; it now says what was recomputed: the published starting values, byte for byte, and every new rare position.

28. 2026-10-29Correction

    The page said one game did not get an OEIS entry of its own. Every game has one: a new entry for the largest game and older entries, extended, for the rest. The count of entries is corrected to match.

29. 2026-10-29Correction

    The page gave the end of the expert's later run on the largest game as his furthest point; his table now goes much further, and the page says so. It also said a from-scratch recomputation gave byte-identical output; it now says what was recomputed: the published starting values, byte for byte, and every new rare position.

30. 2026-10-29Correction

    The page said one game has no OEIS entry of its own. Every game has one: a new entry for the largest game and older entries, extended, for the rest. The count of entries is corrected to match.

31. 2026-10-29Correction

    The page said the correction was submitted together with freshly computed values for the terms around it, which could be read as changing those terms too. It now says the recomputed neighboring values were sent as evidence and that only the corrected term changed. It also gives the time the wrong value stood with its start and end dates.

32. 2026-10-29Correction

    The post said the July search found no earlier count, and the specialist notes said no tabulation was found. A later literature check found one earlier table, for small numbers of tasks, in a paper in the Journal of Artificial Intelligence Research, and a recurrence in a later conference paper. The post now cites both. The OEIS entries are the first for these counts and go far past that table.

33. 2026-10-29Correction

    The proof sketch said a task's number of options depends only on how many tasks sit on the lower levels. It depends on two numbers: how many sit on the level just below, and how many sit on the levels under that. The page now says so. It also says the earlier published table matches these values entry for entry, who did the later literature search, and that the four computations were all written by Claude agents.

34. 2026-10-29Correction

    The page said the table's keeper had since doubled the range for this game. His table now goes much further, with no new rare value and a slightly larger largest value; the page now says so.

35. 2026-10-28Accepted

    A401259, A401447 and A401448 published in the OEIS

    They count shapes built from one more space-filling cell, a 7-faced one, in the three standard ways (fixed, one-sided and free). The number of OEIS entries crediting this work moved from 43 to 46, 22 of them authored.

36. 2026-10-27Correction

    The summary said this check is linked from Terence Tao's database of Erdős problems. The database links the OEIS tree-count sequence, added by my pull request; the summary now says that.

37. 2026-10-27Milestone

    Wording tightened and the sentence on the merged pull request moved to its own page; the page now says the conjecture also covers forests and this check covers trees only, credits the earlier computation to B. Reynolds, and points to the proof claims posted later.

38. 2026-10-26Milestone

    This page records my own work as it stood in July: the conjecture, how far I checked it, and why I could not prove it; the few later additions are dated. The later proof by Roberto Blanco Gómez now has its own page, linked at the end.

39. 2026-10-26Correction

    Credit and accuracy fixes: the page now credits the earlier recurrences of Eggeling and co-authors and Robinson, and states the uses of the count, the Markov-equivalence note and the reach of the programs more exactly. Notes at the unproved pattern point to the proof posted later, a preprint not yet refereed.

40. 2026-10-24Milestone

    A dropped digit sat in an OEIS entry for years until a Claude agent spotted it. What that means for anyone who copies values from reference tables.

41. 2026-10-24Milestone

    Checking an Erdős conjecture on every tree up to a size rules out small counterexamples but does not prove it. What such a check shows and how it is made.

42. 2026-10-22Milestone

    Said exactly how far the second program agreed (up to six cells), relabeled the exact-fraction rebuild as an internal check, and added the third count (the fixed count, already in the OEIS since late August) to the text of this page.

43. 2026-10-22Milestone

    The check box no longer says the count match shows each tree was produced once; it shows none were lost or added overall.

44. 2026-10-22Milestone

    Corrected the explanation of the open question: a finite table can prove that a game repeats (a theorem of Guy and Smith), but not that it never does.

45. 2026-10-22Milestone

    Said exactly what the tables' maintainer reran: the largest game over his old range, two other games over their full new ranges, and a fourth game matching later.

46. 2026-10-22Milestone

    Corrected the explanation of why the open question is open: a finite table can prove that a game repeats (a theorem of Guy and Smith), but it cannot prove that one never does. Added the game's rules and what counts as a rare value.

47. 2026-10-20Accepted

    A401255 through A401258 published in the OEIS

    They count shapes built from two more space-filling cells, an 11-faced one and a 15-faced one. A401257 and A401258 were published on October 19 and A401255 and A401256 on October 20; an outside contributor had computed the counts independently and they matched. The count of OEIS entries on this site moved from 39 to 43.

48. 2026-10-20Accepted

    "Verify Before You Accept: 33 Days of Incident-to-Rule Practice in an Operator-Supervised Agent Hierarchy" accepted as a poster at the 3rd Workshop on Lifelong Agents, AIAS+ 2026

    The paper studies the records of my own team of AI agents: what the lead agent caught by re-running or checking work before accepting it, and what was found wrong later. The workshop meets in San Francisco on November 7, and it is not archival.

49. 2026-10-17Accepted

    A400457 through A400461 published in the OEIS, all five on the same evening

    They count shapes built from the cells of two space-filling solids. The count of OEIS entries on this site moved from 34 to 39.

50. 2026-10-11From others

    A Lean version of the all-forests proof claim for Erdős problem 993 was registered on Palomar and accepted by both of its proof checkers

    The proof is by Wei Li, Kevin Vallier and Tong Zhang (arXiv:2610.07943), revising the claim Tong Zhang posted on September 27. It covers every forest, so every tree, including all the trees my check went through. The paper has not been refereed, and the GitHub database of Erdős problems still records the problem as open.

51. 2026-10-07From others

    Roberto Blanco Gómez posted a proof of the A397711 remainder conjecture as a preprint on Zenodo

    He also added a comment to the OEIS entry pointing to the preprint, which an editor approved on October 12. The preprint has not been refereed.

52. 2026-10-05Milestone

    Both workshop papers posted on Zenodo, free to read

    The final versions of "Rediscover Before You Discover" and "How Far a Wrong Claim Travels Before It Is Caught" are now open access, each with a permanent DOI: 10.5281/zenodo.23154795 and 10.5281/zenodo.23154707.

53. 2026-10-04Milestone

    Registered to present both workshop posters at NeurIPS 2026, at the Atlanta site

    The conference confirmed both presentations and the registration in its invitation letter. NeurIPS 2026 runs across three sites, Atlanta, Sydney and Paris, from December 6 to 13.

54. 2026-10-04Milestone

    An entrance test for an AI-built search program caught a rule that would have thrown away the very solid it was looking for. Then a reviewer's pointer showed the "new" solid had been printed, unnamed, in a 2016 survey.

55. 2026-10-01Milestone

    Added that the solid's shape had already appeared, without a name, in Moritz Schmitt's survey of such solids.

56. 2026-09-30Accepted

    "How Far a Wrong Claim Travels Before It Is Caught" accepted as a poster at AI-Native Academia, NeurIPS 2026

    The paper looks at what the program’s checks caught, which check caught each thing, and how early. The workshop meets in Atlanta on December 12 and 13.

57. 2026-09-28Accepted

    "Rediscover Before You Discover: A Recent Human Find as the Regression Target for an Agentic Geometry Pipeline" accepted as a poster at the MATH-AI workshop, NeurIPS 2026

    The paper argues that an AI discovery pipeline should first re-find a recent human discovery, with almost nothing given to it, before it is allowed to claim anything new. The workshop meets in Atlanta on December 12.

58. 2026-09-28Correction

    A wider search found the Josehedron’s shape type listed, unnamed, in a 2016 survey; the sequences counting its shapes are still new

    The earlier check had compared it with only one table of that survey. Exact recomputation confirmed the match. The three OEIS sequences counting Josehedron shapes are still new.

59. 2026-09-26Sent out

    Sent five new shape-counting sequences to the OEIS in one evening, A400457 through A400461

    They waited for the editors until October 17 and were not counted on this site until then.

60. 2026-09-26Milestone

    A network puzzle: pick as few points as possible so that every point is picked or linked to a picked one. A search I ran with Claude found a smaller set than any published.

61. 2026-09-26Milestone

    When you hand someone results made with AI agents, they need to know what they can check from what you gave them and what they have to take on trust. An editor of a public math database asked me exactly that about a count I had added. The honest answer was "at least this many", not "exactly this many".

62. 2026-09-25Accepted

    A307185 published with its new term, 5,539,121

    Approved after two weeks with four OEIS editors. The entry was already counted for an earlier term, so the count of OEIS entries on this site stays at 34.

63. 2026-09-20Stopped

    Stopped a month-long computer search on the twelfth Fibonacci cube without an answer, and published the attempt

    After 29 days and 22 hours on one core, the solver had not decided whether 54 points can cover the whole network. The known range, 54 to 59, did not change, and the verified set of 59 remains the result. The search, its inputs and its record were published that night and later archived on Zenodo.

64. 2026-09-18From others

    Terence Tao merged the seven links between OEIS sequences and Erdős problems into his public database

    Problems 308, 634, 939, 942, 943, 1140 and 1161. The second merged contribution from this program, after problem 993 in August. A merged link records the connection; it is not an endorsement.

65. 2026-09-16Correction

    The Erdős tree check's title and claim now name the independent-set sequence (the counts of ways to pick dots with no line between them), the one the problem is about.

66. 2026-09-14Correction

    Replaced the word "certificates" with "canonical-form lists" in the A307185 entry and its archive, after an OEIS editor asked what it meant

    The graph software used here calls its labeling strings certificates, but to a mathematician the word means a proof. The archive now says plainly what a reader can check from the files alone and what needs the program run again.

67. 2026-09-12Accepted

    A258205 published in the OEIS up to size 32

    Thirteen new terms, credited by the OEIS. Its companion sequence A258204 was published on September 8.

68. 2026-09-11Sent out

    Sent the new term of A307185 (a count of phylogenetic network shapes) to the OEIS

    An OEIS editor worked on it the next day, and it then waited in the editors’ queue.

69. 2026-09-11Milestone

    I keep a record of every error a check caught in my math work with AI agents: who made it and what caught it. Two of them changed how I check everything.

70. 2026-09-07Milestone

    Published two new public code-and-data repositories, each archived on Zenodo

71. 2026-09-06Accepted

    Improved the published upper bound for the domination number of the twelfth Fibonacci cube from 61 to 59, now stated in OEIS A291295

    An explicit set of 59 guards, checked by three separate programs, was attached to the entry. The published lower bound of 54 is unchanged.

72. 2026-09-05Correction

    Withdrew the claim that the space-filling catalog was the only machine-readable list of its kind

    The claim was made before a full search. A 2026 database by Ed Pegg appears to cover the same 2016 survey and is now cited as related work. The catalog is described only as not matched against the records checked as of a given date.

73. 2026-09-05Sent out

    Proposed seven more links between OEIS sequences and Erdős problems in the public database

    One-line links for problems 308, 634, 939, 942, 943, 1140 and 1161.

74. 2026-09-05Stopped

    Both record attempts stopped under their own rules, without beating either record

    The Ramsey search reached 276 against a known bound of 381. The code search reached 49, equal to the known best, and ruled out 50 within its search.

75. 2026-09-04Result

    The second stage of the space-filling search certified 165 more cells (14 from the tetragonal crystal family, 151 from the hexagonal family); none named, none matched against the records checked

    Before trusting anything, each batch reproduced every printed row of a 2016 survey for its crystal family: 1,476 tetragonal rows and 958 trigonal and hexagonal rows, the latter typed in a second time by a second reader who had not seen the first copy, with no discrepancies.

76. 2026-09-04Result

    Sorted all 165 certified cells by what a small nudge does to their shape type, under rules written in advance: 115 keep their type (open), 41 sit on a boundary where it changes on both sides (wall), 9 change it on one side only (one-sided)

    Each cell was nudged a little in every direction tested and its shape type recomputed. Open means the type survived every nudge; wall means that in some direction it changed whichever way the cell was nudged; one-sided means it changed in one direction but not the opposite one. The quick labels guessed earlier disagreed with the computed answer on 35 of 149 cells, in both directions, and were dropped.

77. 2026-09-04Milestone

    Built a machine-readable catalog of 1,583 exactly confirmed space-filling cell types, 177 of them certified, and matched it against all 3,315 printed rows of the 2016 survey

    The three hand-entered tables add up to the survey’s own stated totals, and a separate recount sharing no code found zero failures. No DOI yet.

78. 2026-09-04Result

    Computed two polyhex counting sequences, which had stood at size 19 since July, up to size 32, four independent ways

    A258205(32) = 808,490 and A258204(32) = 750,157,868,462. The method: count only the mirror-symmetric shapes, then get the rest by subtraction from a published column.

79. 2026-09-04Correction

    Confirmed the cause of the August polyhex discrepancy, from notes the project had written in July

    Two programs had agreed on the same wrong values because they shared one step that identifies a shape by its corner points, which loses information for larger shapes. A different explanation recorded in August was withdrawn.

80. 2026-09-04Started

    Launched two attempts at known records under rules written in advance: one on an error-correcting code over five symbols, one on a four-color Ramsey number

    Preparation showed the first target could not be checked on this machine in under about five CPU-years, so it was switched to a neighboring one that could be. Three exact checkers agreed on 208 published codes before launch, and test runs caught three bugs in the search code first.

81. 2026-09-03From others

    Flammenkamp updated his octal tables, taking all twelve games from July further; the July credit stays on his page

    So every octal record on this site is dated July 2026, and none is described as current.

82. 2026-09-03Correction

    Corrected the running OEIS count from 34 to 33 after a recount found one entry counted twice

    A202182 had been counted as both an extension and a correction. This site had already shown 33 since August 29.

83. 2026-09-03Result

    Computed a new term, 5,539,121, for sequence A307185, with two separately written programs agreeing exactly; not yet sent to the OEIS

    Both programs first reproduced every published smaller term, down to the individual building blocks.

84. 2026-09-01Correction

    Caught a wrong method name in the benzenoid census paper by checking it against the code

    The paper described a backup method for larger cases, but no case was large enough to need it, and the paper had given it the wrong name. That was confirmed against the code two days later, and the paper was rebuilt around how its measures behave as molecules grow.

85. 2026-09-01Milestone

    A search for new space-filling solids ended with seven cells that passed every exact check, two of them named the Satchelhedron and the Ordenhedron

    None of the seven matched any shape in the reference records checked as of that date. Passing the checks means each one tiles space as claimed; it does not by itself mean the shape is new.

86. 2026-08-27Accepted

    A398957 published, completing the set of Josehedron counting sequences in the OEIS

    The editor asked for the description of the process to come out and the checked counting program to be uploaded as a file; both were done before publication.

87. 2026-08-26From others

    The census authors asked to see the write-up, and it went to them with the section on errors in their tables marked for comment

88. 2026-08-22Milestone

    Archived the code and data for five projects on Zenodo, each with a permanent DOI

    These archives are the public record behind the papers.

89. 2026-08-17From others

    Sent a courtesy note to the authors of the benzenoid census about six errors found in their paper

    The note said I had found six small errors in the published tables and a figure caption, with corrected values, and offered the manuscript. No reply about the errors themselves is on record.

90. 2026-08-16Correction

    Found a sixth error in the published benzenoid census, this time by reading rather than computing

    Five errors had already turned up when the published tables were recomputed exactly. The sixth is a place where the paper’s text contradicts its own table.

91. 2026-08-15Accepted

    Published counts of shapes built from Josehedron cells, the first in the OEIS, as A398958 and A398959

    One sequence counts a shape and its mirror image as different; the other counts them as the same.

92. 2026-08-15Result

    Extended the breakdown tables of a published benzenoid census by two sizes, with totals matching the published sequence exactly

    The OEIS (A018190) already listed the total number of benzenoids at these sizes; what is new is the census’s breakdown of each total. The run could not report any table until its totals matched the published values, and both did.

93. 2026-08-09Milestone

    Posted the result for every tree with 32 vertices on the Erdős problem's forum

    With it, every tree with up to 32 vertices had been checked, about 165 billion more trees than the earlier checks covered. Proof claims for the problem were listed later, on September 21 and 27, and neither cites the check; a Lean version of the second, which covers every forest, was accepted by the Palomar registry's two proof checkers in October.

94. 2026-08-08Correction

    Corrected the OEIS fullerene table A057210, whose 2006 data file predated a bug fix in the program that made it

    Thirty-three values were replaced and 28 of them had been wrong. An editor published the fix and a second editor reviewed it.

95. 2026-08-07Accepted

    The OEIS table of largest Sidon sets, A143824, published extended from 500 to 586

    All 501 terms already published were reproduced first. The table can go no further until a longer optimal Golomb ruler is known.

96. 2026-08-06Milestone

    Surveyed the field and prepared fifteen possible problems, checking each one against its sources

    Nothing was computed: only notes, checked citations, known values to test against, and stopping rules written in advance. Going back to the raw sources caught three tables of numbers an AI agent had invented.

97. 2026-08-06Accepted

    Terence Tao merged a contribution into his public database of Erdős problems

    It adds the OEIS tree-count sequence, A000055, to problem 993's entry; the pull request's description cites my check, which took the check of every tree to 32 vertices. The merge adds a link to a database; it does not endorse a proof.

98. 2026-08-05Accepted

    The last three of eleven extended octal-game sequences were published on the same day, completing that work

99. 2026-08-05Decision

    Set aside a well-known open problem after finding others already far ahead

    Earlier work had already searched far past anything a simple approach could reach, one claimed proof had been publicly refuted, and four people were listed as working on it. No computing time was spent.

100. 2026-08-04Result

     The tree check finished at the next size and matched the published count exactly

     No counterexample in any of sixteen separately run pieces.

101. 2026-08-04Sent out

     Proposed a link between one problem in the public database of Erdős problems and its OEIS sequence

     The counts were checked three times against the published data, and the proposal disclosed AI assistance as that collection asks.

102. 2026-08-03Result

     Extended a Sidon-set sequence (sets of numbers whose pairwise sums are all different) and checked every value against the published terms

     No mismatches with the published terms, and the underlying values were checked a second, independent way.

103. 2026-08-03Decision

     Started a dated log of the program in which lines are added but never edited

     When something changes, a new line records it. Dead ends stay listed, marked as dead ends.

104. 2026-08-02Result

     Checked every tree with 30 and 31 vertices for an open Erdős problem, past the earlier limit of 29, and posted the results on the problem's forum

     Others had checked every tree with up to 29 vertices earlier in 2026: the forum user JakeMallen reported it on the problem's forum in January, and Brett Reynolds published code and a preprint in March. That work is credited alongside, and the smaller sizes rest on it.

105. 2026-07-22From others

     The extended octal data appeared on Flammenkamp’s reference page, with credit by name

     He had already reviewed the program and reran parts of the computation with his own programs. His page lists eleven games; the work covered twelve, and statements about the page use his count.

106. 2026-07-17Milestone

     When I rechecked everything an AI model produced during a switch I had not been told about, every computed number came out the same when it was recomputed. The errors were in the wording around the numbers and in how some checks had been done; one real finding had also gone unnoticed.

107. 2026-07-16Correction

     A review of two and a half days of work found a discovery that an earlier summary of the results had missed

     One new rare value in a game whose run had been recorded as finding none.

108. 2026-07-16Correction

     The same review found that an overnight check against the OEIS had missed matches, because sequences can start counting at different positions

     All ten games turned out to match existing entries, so the work was extending sequences rather than creating new ones. This was corrected before anything was sent.

109. 2026-07-16Milestone

     Every run first had to reproduce the expert's published tables exactly. That let him check the new work on his own machines instead of taking my word for it.

110. 2026-07-16Result

     In July I pushed the published tables for a set of unsolved take-away games two to four times further. In four of the games, new heap sizes with rare values turned up.

111. 2026-07-16Milestone

     WSJF is provably the best order for independent work. One dependency, or SAFe's scoring sheet, can make it put the wrong item first.

112. 2026-07-15Stopped

     A deliberately skeptical check of the WSJF paper confirmed that its central theorem was already known, and the paper was set aside

     Its central result was a theorem from 1956.

113. 2026-07-15From others

     Flammenkamp checked the octal program himself and confirmed the results

114. 2026-07-13Correction

     Corrected a dropped digit that had sat in a published sequence for fourteen years

     The correct value was worked out by two routes that do not depend on each other before the correction was sent. The sequence’s original author replied: "great, thanks!"

115. 2026-07-13From others

     The OEIS board updated its policy on AI so that it no longer depends on which tools a contributor uses

116. 2026-07-13Correction

     Caught a stray note in a draft before it was sent to the OEIS

     It never reached the database. Reading the actual text before sending, rather than trusting it, is what caught it.

117. 2026-07-13Milestone

     Gathered the review records from eight projects into one archive

     The record now lives in one place instead of being scattered across projects.

118. 2026-07-12From others

     An OEIS editor asked for three data-file uploads to be redone under a formatting rule, and was right

     The rule: a full data file must not simply repeat the short list of terms shown on the entry. The files were withdrawn and redone.

119. 2026-07-11Accepted

     Three new sequences published in the OEIS, with their full data files uploaded the same day

120. 2026-07-08From others

     Sent the octal results to Achim Flammenkamp, who held the earlier record and maintains the reference tables

121. 2026-07-05Accepted

     The added terms were published in the OEIS about an hour after they were sent

122. 2026-07-05Result

     Computed the octal game 0.007 four times further than the record that had stood since 2021

     Seven hours on one core of an Apple M3 Max laptop. This was the record in July; the tables have moved on since, as the entry explains.

123. 2026-07-05Sent out

     Submitted OEIS A397711, the count of dependency charts in which each task waits on at most two others, with a conjecture about its remainders

     The entry states the remainder pattern as a conjecture, checked by computer up to 100 tasks. An OEIS editor approved it on July 12.

124. 2026-07-05Result

     Showed that copies of the sphenoid, an eleven-faced cell that fills space, can fit together in only one way when every cell is attached by the same rule, then counted the shapes built from it

     I found no earlier counts, and no description in print of exactly how the cells fit together. The counts were later published as OEIS A397708.

125. 2026-07-05Correction

     A check of a study of television transcripts against its original sources found errors in the source data; the analysis itself held

     Nine episodes had the wrong transcript attached, and a third of one series had been cut off by a spreadsheet cell limit. Comparing against the original source is what caught it.

126. 2026-07-05Correction

     Re-ran the whole television-transcript analysis on the corrected data; every finding held

127. 2026-07-05Milestone

     Shapes built by gluing copies of two space-filling solids face to face: more counts for the bisymmetric hendecahedron, and the first I could find for the cell of the sphenoid honeycomb.

128. 2026-07-03Started

     Started the program: look for small results that can be computed in under an hour and then checked

     Three lines of work began the same day: continued exponentials (numbers built from towers of exponents), polyforms, and octal games.

129. 2026-07-03Stopped

     Dropped the continued-exponential work the same day, after finding the result already published

     The main result was already stated in the abstract of a 1996 paper. Checking for earlier work before computing anything is why this cost an afternoon instead of a week.

130. 2026-07-03Result

     Computed four new terms of a polyform counting sequence that nobody had extended since it was created

131. 2026-07-03Result

     Computed one of those terms again with a second, separately written program before sending it in

     A different program in a different language gave exactly the same number.