The 28th CGPM: three time drafts, three days in Versailles
One meeting decides three questions about time, and one of them is lunar. This page holds the agenda, the verbatim subject of each draft, and what each outcome would and would not settle.
Key facts
- The 28th CGPM is convened for 13–15 October 2026 at the Palais des Congrès de Versailles, Yvelines, France. [1]Official
- Three drafts on its provisional agenda concern time: B is item 15, C is item 16 and D is item 17. [1]Official
- Votes on all resolutions are item 25 of the provisional agenda, after proposals by delegates at item 24. [1]Official
- Decision CIPM/115-5 endorses an updated Draft Resolution C that fixes 20 May 2027 as the implementation date of the new maximum value for the difference UT1 − UTC. [2]Official
- Decision CIPM/115-14 states that there must be a unique lunar reference time scale, even if there may be several realizations. [2]Official
Last checked: . On that day we re-read the convocation of the conference and the page of CIPM decisions behind the three drafts below. [1]Official
The General Conference on Weights and Measures meets every four years. Its 28th meeting has six lettered draft resolutions in front of it, and three of them are about time[1]Official. This page follows those three: B on the second, C on the continuity of Coordinated Universal Time, and D on a lunar reference time scale.
When and where
The conference runs 13–15 October 2026 at the Palais des Congrès de Versailles, 10 rue de la Chancellerie, 78000 Versailles, Yvelines, France[1]Official. The meeting is convened in a hybrid format, with both in-person and remote participation. It is chaired by Prof. Françoise Combes, President of the Académie des sciences (France)[1]Official.
Opening day is : in 23 days from , the day this page was built. The count is our own arithmetic — whole days between two dates, nothing else[3].
| Date | Day | What the convocation places here | Source |
|---|---|---|---|
| Day one | The conference opens. Credentials are presented, the list of delegates entitled to vote is established, the agenda is approved, and the Special Procedure for hybrid conduct is considered in the opening session. | [1]Official | |
| Day two | Business continues through the numbered agenda. From item 11 onward the convocation gives no order and no times: it says they will be set out in a detailed timetable published separately. | [1]Official | |
| Day three | The meeting closes on this date. Votes on all resolutions are item 25 of the provisional agenda, after proposals by delegates at item 24, and the convocation does not say which day item 25 falls on. | [1]Official |
The convocation does not attach an hour to any of the numbered items after item 10. It says that their order and timings will be given in a detailed timetable published separately, and we will not guess at one [1]Official.
The three drafts we follow
The subject of each draft below is quoted word for word from the convocation of 13 January 2026, the document member states received. The fourth column is the status we can confirm from a primary source, not a forecast.
| Agenda item | Draft | Subject, verbatim from the convocation | Status | Source |
|---|---|---|---|---|
| 15 | B |
| not yet voted | [1]Official |
| 16 | C |
| not yet voted | [1]Official |
| 17 | D |
| not yet voted | [1]Official |
What each outcome would change
Two things can happen to each draft, and both are worth setting out before the conference rather than after it. The paragraphs below are conditional on purpose: they describe consequences, and they say nothing about which consequence will follow.
Draft Resolution B — agenda item 15
The SI second itself — the unit every other time scale on this page is built from.
- If the conference agrees the text
- The conference would put the redefinition of the second on its formal path. The convocation lists the title of the draft and nothing else, so its text, and any date inside it, are not something we can report from a primary source yet. [1]Official
- If it does not
- The definition of the second in force today would be unchanged, and nothing else on this list would move with it: B is about the unit, not about UTC and not about a lunar scale. [1]Official
Draft Resolution C — agenda item 16
The continuity of UTC, and with it the future of leap seconds.
- If the conference agrees the text
- Decision CIPM/115-5 endorses an updated text that fixes 20 May 2027 as the implementation date of the new maximum value for the difference UT1 − UTC. If the conference agrees that text, that is the date the change is made. The numerical value of the new maximum is not given in the published decisions. [2]Official
- If it does not
- Resolution 4 of the 27th CGPM stands on its own: it already decides that the maximum value for the difference UT1 − UTC will be increased in, or before, 2035. Without a new text the outer limit of 2035 remains and no earlier implementation date is fixed. [4]Official
Draft Resolution D — agenda item 17
Whether an international lunar reference time scale is defined at all, and how it is tied to UTC.
- If the conference agrees the text
- The conference would have a definition on the table and a stated requirement of traceability to UTC. It would still not name an organization to compute the scale, and it would not start one: decision CIPM/115-14 says only that there must be a unique lunar reference time scale, even if there may be several realizations. [2]Official
- If it does not
- The IAU definitions of 2024 and the national deadlines would stay where they are. The United States memorandum of 2 April 2024 asks NASA for a finalized lunar timing implementation strategy no later than 31 December 2026, whatever happens at Versailles. [5]Official
The outer limit behind Draft Resolution C was set four years earlier: Resolution 4 of the 27th CGPM decides that the maximum value for the difference UT1 − UTC will be increased in, or before, 2035 [4]Official. That resolution does not print a number for the new maximum, and neither do the CIPM decisions[2]Official. What is settled and what is not is separated line by line in leap seconds and the October vote.
National deadlines are not on this agenda and do not move with it. A United States memorandum of 2 April 2024 asks NASA for a finalized lunar timing strategy no later than 31 December 2026 [5]Official.
Where we will publish the outcome
Nothing on this site moves until a primary source says it moved. When the conference has taken each item, the result appears in these four places, in this order: the tracker row first, then the timeline entry, then an article, then the email brief.
- Lunar time standard tracker — the board and its dated changelog
- Timeline — one entry per document that lands
- News — the article that explains what changed
- Draft Resolution D in full
- Lunar Time Brief — the email that follows the board
The status column above is a field in our data, not a sentence in this page. It changes in one edit, and the change is logged in the changelog on the tracker with the document that justified it.
How we keep this page
This page carries no forecast and no countdown written by hand. The one number that changes on its own is computed when the page is built, from the build date, and it is labelled as such.
Where a document does not answer a question, the page says so instead of filling the gap. The convocation gives no timetable for items 11 onward, so this page gives none. The published decisions give no numerical value for the new maximum of UT1 − UTC, so this page gives none [2]Official.
The reading behind each of the three drafts is collected in the lunar time news hub, and the chain of documents that leads to this meeting is on the timeline.
Sources
Last verified