Zulip Chat Archive
Stream: mathlib4
Topic: Naming of `Simplex.points`
Jovan Gerbscheid (Apr 30 2025 at 09:45):
I don't like the name Affine.Simplex.points
, because in practice this means that to refer to vertex i
of a simplex s
, we write s.points i
. And I get really confused by the plural points
. (Compare Affine.Simplex.faceOpposite
, which isn't called facesOpposite
). I think this should be renamed to point
or vertex
. And points
or vertices
could be defined as s.points : Set P := Set.range s.point
.
Jovan Gerbscheid (Apr 30 2025 at 09:52):
An orthogonal quesiton is why we use points
instead of the more common vertices
.
Joseph Myers (Apr 30 2025 at 23:51):
Whatever you call it, having a coercion to function would allow you to write s i
and use the name a lot less.
I suspect the affine span of the set of vertices is used a lot more than the set of vertices is used directly (and so has a stronger case for having its own definition and associated API for that definition).
Eric Wieser (May 01 2025 at 00:18):
I agree that singular would be clearer here
Last updated: May 02 2025 at 03:31 UTC