FAQ & Change Log

Frequently Asked Questions

Rarity Tier Key

Each tier boundary comes from an exhaustive scan of consecutive days, not a guess — these are genuine order-of-magnitude breaks in the real distribution of scores across history.

"The Witness" is deliberately not framed as "extremely rare" — after the correction below, score 12 is currently unique across the entire searchable historical record: 2 April 33 AD, the evening of the Resurrection, and nothing else. That is a difference in kind, not just in degree, from every tier below it.

The Venus Criterion Was Too Broad

A full historical sweep (BC 4000–AD 3000) turned up five dates scoring a perfect 12, not one. Four of the five landed on the Venus cycle's opening day of one of seven different phases — Falcon, Light, Lotus, Serpent, Temple, and Wind were all being credited as Major, with no distinction between them.

Only one of Venus's nine phases carries the Resurrection typology this whole project is built around: Reed, the morning the Morning Star reappears after being hidden behind the sun — the phase this project's own reference material already describes as the Resurrection phase. The other six openings were receiving the same credit despite carrying no equivalent significance.

The fix: only Reed's opening days remain Major. The other six phases were demoted to Minor rather than dropped to Normal entirely — they're still genuinely notable Venus configurations, just not equal to the one phase this calendar treats as Christologically central. The four affected dates now read as Exceptional (score 11) rather than The Witness (score 12); none of them were disqualified by accident or overcorrection — each still carries real, substantial convergence.

Venus's overall Major frequency dropped from 2.2% of days to 0.27% — an eightfold tightening, matching the arithmetic of restricting from seven eligible phases to one.

The Priestly Course Criterion Was Too Broad

Before the Venus fix above, an audit of all six tracks' independent Major-frequency rates found one clear outlier:

Priestly Courses was carrying roughly double the next-highest track's rate. The cause: six of the 24 priestly courses were classified Major on any day of their seven-day week, rather than only on the week's actual sabbath — the specific day (noon Saturday to noon Sunday) this project's own reference material already identifies as each week's holiest.

The fix restricted Major (and Minor, for the other 18 courses) to week-day 1 exclusively. Priestly's Major rate fell to 8.2%, bringing it in line with the other five tracks rather than standing apart from them. All five of the perfect-12 dates then known were re-checked against the corrected rule and every one of them already happened to land on week-day 1 — the tightening didn't retroactively disqualify anything, it simply closed a gap that other, unrelated dates had been exploiting.

Calendar View: Sunset–Midnight Slice Navigating to the Wrong Day

In Enochian calendar view, each day cell's first slice (Sunset–Midnight) is deliberately sourced from the previous Gregorian day's own evening — because an Enoch day begins at the prior evening's sunset. The underlying data was always correct, but clicking that slice navigated to the cell's own date instead of the source date it was actually displaying, landing one day later than intended. Fixed by having the click target follow the same source-date tag the tooltip already used.

Search Functions Retired the Legacy Night/Day Binary

Every search-family tool — Search, Range Analysis, Fingerprint, Anniversary, and the Next/Previous event finder — originally scored each candidate day using a coarse legacy toggle: a blanket "Night" or "Day" state that wasn't tied to any real clock time or location. It was an artifact of an earlier computation model, superseded once Calendar View introduced genuine four-period (Midnight/Morning/Afternoon/Evening) slice-level accuracy.

All search functions now sample the same four real, Jerusalem-anchored periods Calendar View uses, computed fresh for each specific date using a direct port of the same solar-position formula the frontend already relied on — verified numerically identical to full floating-point precision. Every search result now also carries the literal boundary values that produced its score, so clicking through reproduces the exact number shown rather than a different one computed under a different time assumption.

Solar Eclipse Detection Was Far Too Liberal

A user reported today's calculator reading as a solar eclipse when it plainly wasn't. A real eclipse requires two things at once: the moon near conjunction with the sun, and the moon near one of its two orbital nodes (a small ecliptic latitude). The calculation only checked the first condition — but at conjunction, those two numbers are nearly identical, so that check alone flags the majority of all new moons, eclipse or not. The equivalent check for lunar eclipses had always correctly included both conditions; the solar-eclipse case appears to have simply lost its second condition somewhere along the way.

Measured against a real astronomical ephemeris and the exact geometric definition of a partial eclipse, across every new moon from 1900–2050: the old calculation produced 868 false positives — 57% of every non-eclipse new moon in that 150-year span, roughly six phantom eclipses a year.

The fix restores the missing condition, calibrated so every real eclipse in the 150-year test range still clears it with room to spare, while removing 85% of the false positives. The 134 that remain are genuine borderline cases — near-misses this simplified model of the moon's motion can't fully resolve without a substantially more complex lunar calculation.

One existing devotional date-selection turned out to have been influenced by this: a candidate date for Isaac's binding at Moriah had scored a point higher than it should have, on the strength of a solar eclipse that, per the correction and independent verification, never actually happened. Re-examined under the fix, the date's other genuine convergences still held up — but a different candidate age now wins the comparison. The site's search index and every catalogued event's stored score have been rebuilt against the correction.

The Zodiac Disc Was Computing Precession From the Wrong Year

The interactive disc rotates its outer, star-based ring relative to the seasons to represent roughly 26,000 years of axial precession — but that rotation was quietly being calculated from the Enoch calendar's own internal year-count rather than the real calendar year. It was a silent, universal mix-up: every single date, on every load, was affected — it simply went unnoticed until a specific historical date's star sign visibly landed in the wrong place on the ring.

Fixed by making sure the real year and era are available wherever the disc's rotation is calculated, so it now anchors to the same date the rest of the calculator already shows.

Rarity Search Results Loaded the Wrong Time of Day

Clicking a Rarity Search result populated the date but not the specific time slice (Midnight/Morning/Afternoon/Evening) that had actually produced the shown score, so a result reading "Exceptional, Evening" could load at an unrelated time and compute a different, lower score entirely. Fixed so results now load at the exact time that produced them.

New This Week

• Historical event places, and some times — 517 of 528 catalogued events now carry a real (or, where the record only allows it, deliberately generalized) geographic location, and a handful carry a specific or approximate time of day where one is genuinely documented. Clicking through to an event now loads its actual place and time, not just its date.
• Tap any of the six calendar cards for more — each of the six cards on the main calculator now opens the same detailed reading the disc shows for that track (previously only the Star Wheel card had this, via hover). Venus's card now also opens with a plain, literal line about its actual current motion — morning star or evening star, rising or descending — ahead of the symbolic reading.
• Zodiac ring redesigned around its real, interlocking boundaries — the sidereal zodiac isn't a simple 12-way equal division; three of its true constellation borders (Libra/Scorpio, Capricorn/Aquarius, Pisces/Aries) cross the ecliptic more than once. The ring now represents this properly with two touching half-height rings at each of those three points, rather than either ignoring the overlap or drawing it as unreadable slivers. Each sign is now marked with its own zodiac glyph rather than spelled-out text.

Recent Additions

• Rarity tiers — replacing the bare 0–12 score with a named, star-rated tier calibrated against measured frequency (see the key above).
• Precision caution — dates before roughly 500 BC now carry a note that the underlying solar-position formula's accuracy narrows with distance from the present; ancient-date findings aren't invalidated, but their timing carries more uncertainty than recent centuries.
• Nearest historical event, surfaced automatically — any date scoring Rare or above now shows its closest catalogued historical event inline, with one click through to the full before/after listing.
• CSV export — Search, Range Analysis, and Fingerprint results can now be exported directly for offline analysis.
• Shareable permalinks — every calculated date can be copied as a direct link (including a custom location, if one is set) via the Share button.