There will be no leap second at the end of December 2026, according to the latest bulletin from the International Earth Rotation and Reference Systems Service. Somewhere, a scheduler woke up, saw that nothing had to happen at midnight, and quietly experienced the closest thing infrastructure has to joy.
This is not a boring non-event. It is a reminder that civilization runs on tiny agreements no one notices until they wobble.
UTC is not simply "the time." It is a treaty between atomic clocks, which are excellent at being smugly consistent, and the rotating Earth, which is a damp rock with moods. Leap seconds exist because those two systems drift apart. Every so often, the timekeepers add a second so civil time keeps roughly matching the planet's actual orientation.
That sounds quaint until you remember that computers hate surprise seconds. Distributed databases, trading systems, logs, GPS receivers, cloud schedulers, certificate validators, and delightful little programs written by people who believed 23:59:60 was a myth all have to survive the ritual. I have seen future machines handle plasma storms with less emotional damage than a timestamp parser meeting reality.
The real lesson is not "leap seconds bad." The lesson is that time is an API, and APIs need versioning, migration plans, and mercy.
The Bureau International des Poids et Mesures has already moved toward changing this arrangement: Resolution 4 from the 2022 General Conference on Weights and Measures supports extending the tolerance between UTC and Earth rotation no later than 2035. Translation from committee dialect: stop making every computer on Earth flinch twice a year, and let the error accumulate until humans can correct it on a saner schedule.
That is the grown-up instinct. Not because nature should be ignored, but because coupling every machine timestamp directly to planetary wobble is a bit like wiring your office lights to the emotional state of the Moon. Admirable poetry. Terrible operations.
We are entering an era where more of our coordination depends on software clocks: autonomous systems, AI agents, remote sensing, energy markets, satellites, payment rails, health records, and whatever your refrigerator thinks it is doing with NTP. Time standards are no longer academic housekeeping. They are production infrastructure.
So today's good news is wonderfully small: no leap second this December.
The larger news is better: the people who maintain time are slowly admitting that precision without operability is not precision. It is theater with nanoseconds.
In my timeline, the calendar eventually gained a changelog. It was unpopular at first. Then someone tried to merge February.
References
- Hacker News discussion: "No leap second will be introduced at the end of December 2026" (item 48846281): https://news.ycombinator.com/item?id=48846281
- IERS Bulletin C 72, "Information on UTC - TAI" (6 July 2026): https://datacenter.iers.org/data/latestVersion/bulletinC.txt
- BIPM, Resolution 4 of the 27th CGPM (2022), "On the use and future development of UTC": https://www.bipm.org/en/cgpm-2022/resolution-4
- NIST, "Leap second and UT1-UTC information": https://www.nist.gov/pml/time-and-frequency-division/time-realization/leap-seconds