My rough notes for this meeting have been posted to the WG21 wiki (here) and the related github tracking issues updated (CWG 2843, P3717R0; there is no github tracker issue for P3727 yet).

Tom.

SG16 will hold a meeting on Wednesday, June 11th, at 19:30 UTC (timezone conversion).

If you need a .ics file to import into your calendar, you can download it here.

The agenda follows.

• CWG2843: Undated reference to Unicode makes C++ a moving target.
• D3727R0: Update Annex E onto Unicode 15.1.
• P3717R0: Update Annex E onto Unicode 16.
  

SG16 previously reviewed CWG2843 during the 2024-01-10, 2024-02-07, and 2024-02-21 SG16 meetings. Those meetings culminated in consensus to specify Unicode 15.1 as the minimum version for C++26 (and C++23 as a DR) with an allowance for use of a more recent Unicode version in conforming implementations. EWG confirmed this direction during its 2024-03 in-person meeting in Tokyo. Since then, the issue has been awaiting proposed wording that includes an update to the non-normative content of annex E. Steve recently published P3717R0 with the minimal wording needed to update annex E, but with additional scope to bump the minimum Unicode version to 16. Push back on the additional Unicode version change lead Jens to update the proposed resolution in CWG2843 (see the "Additional notes (May, 2025)" section) to apply the minimal update to annex E. Meanwhile, Steve authored D3727R0 which retains the Unicode 15.1 direction already approved by EWG, but with additional suggested updates for annex E that go beyond the minimum strictly necessary to resolve CWG2843; see related discussion on the SG16 mailing list and in particular, Robin Leroy's comments that motivated the additional changes. The goal of this meeting will be to:

1. Determine the minimum Unicode version to recommend as the normative reference in the C++ standard with acknowledgment of the following:
1. EWG has already approved Unicode 15.1 as a DR for C++23.
2. A recommendation to move to Unicode 16 will require an audit to address impact from the rebase, approval from EWG, and, presumably, divergence from C++23 since updating C++23 to depend on a standard that had not yet been published when C++23 was published would be ill-advised. Note that no features added to C++26 depend on Unicode 16.
   
2. Determine the scope of changes to make to annex E:
1. The minimal change needed to resolve CWG2843 (the removal of [uaxid.def.rfmt] as in the May 2025 proposed resolution in CWG2843), or
2. The larger set of changes as proposed in D3727R0 (possibly with additional refinement per discussion on the SG16 mailing list and any further discussion).
   

I haven't scheduled any other papers or topics for this meeting so as to maximize our chances of reaching a final consensus on this issue.

Steve, Robin, and Jens: it would be great if the three of you could collaborate to reach agreement on a unified proposal prior to this meeting (I recognize time is short to do so).

Tom.