Seen in Tokyo in SG1:

Poll: tagged pointers are a valuable facility that we want to see in C++

Result: unanimous consent

Poll: the design of tagged pointers should make an effort to maximize the chances
of being able to use the high order bits

Result: consensus against

P3125R1 constexpr pointer tagging (Hana Dusíková)

We want tagged pointer in constexpr
SF F N A SA
3 6 4 0 0
Unanimous consent

We want tagged pointer schemas (multiple tagging strategies controlled by user,
incl. adopting tagging that pre-exists and incl. implementation-defined tagging)
SF F N A SA
0 2 4 4 2
No consensus

We want tagged pointer to model a pointer (so you don't need to untag to use as a
pointer or iterator)(this does not mean it implicitly converts to pointer)
SF F N A SA
0 3 5 4 0
No consensus