I would describe why, and maybe say a future story could address it based on consistent use. First, the effort to change pod spec would be large (impact). Second, currently our workload controllers are inconsistent and we prioritize consistent of experience over desirable future state (consistency). Third, stateful sets are just different enough from daemonsets and deployments that real world use of minReadySeconds for stateful sets might influence any future design or point in a more appropriate direction (completeness)

I tried explaining second point in the motivation section. I will add first. Can you expand a bit on completeness(third point) like why do you think the real world use of minReadySeconds for sts might point in a more appropriate direction? I can add that in the use-case section.

Link to that k/k issue where this was discussed, for reference.

Be more specific about the risk of a bug, namely that we either:

• mark a pod as ready too soon
• mark as ready too late.
  and what that might potentially mean.

For consistency with other controllers this should not be a pointer, especially that 0 and nil have the same meaning.

Ya, I was thinking about the same but in the apiconventions, it was mentioned that:

• ensure the field is entirely absent from API responses when empty (optional fields should be pointers, anyway) which made me think I should use a pointer. It also helps in setting default values easily as we can distinguish between being set and the default value.

This was also mentioned in the api conventions doc.

Do use pointers to scalars when you need to distinguish between an unset value and an automatic zero value. For example, PodSpec.TerminationGracePeriodSeconds is defined as *int64 the go type definition. A zero value means 0 seconds, and a nil value asks the system to pick a default.

I think the above is the case for us too say if we choose to a non-zero default value in future.

We spoke with @smarterclayton on slack, and the agreement is that we'll go with non-pointer version.

Ditto.