Zulip Chat Archive

Stream: Is there code for X?

Topic: non-submultiplicative normed ring?


Monica Omar (Jul 21 2025 at 13:35):

Is there code for an inner product space that’s also an algebra, but not necessarily that ‖x * y‖ ≤ ‖x‖ * ‖y‖?
So docs#NormedRing but without the norm_mul_le property?

Eric Wieser (Jul 21 2025 at 13:46):

No, and I think probably the plan is to achieve this via changing the spelling of [NormedRing R] to [Ring R] [NormedRing R]

Eric Wieser (Jul 21 2025 at 13:46):

With that change, what you want would be [Ring R] [NormedAddCommGroup R]

Monica Omar (Jul 21 2025 at 13:48):

Eric Wieser said:

With that change, what you want would be [Ring R] [NormedAddCommGroup R]

Yes, exactly this. But this would create non-defeq diamonds.

Yaël Dillies (Jul 21 2025 at 13:54):

Not if NormedAddCommGroup R doesn't extend AddCommGroup R

Monica Omar (Jul 21 2025 at 13:55):

Right. So is the plan to refactor NormedAddCommGroup?

Monica Omar (Jul 21 2025 at 13:58):

perhaps to take in AddCommGroup and extend Norm and MetricSpace?

Eric Wieser (Jul 21 2025 at 14:00):

Monica Omar said:

Right. So is the plan to refactor NormedAddCommGroup?

More of a concept of a plan

Eric Wieser (Jul 21 2025 at 14:00):

#24040 implements this idea


Last updated: Dec 20 2025 at 21:32 UTC