Matt Butcher

Is this something we want to do before the Alpha 1 release of the spec?

The definition for `1` was always intended. Non-containerized workloads were supposed to be captured in extended workload parameters. The problem with treating them as anything else is that we cannot...

Do we need this? Or can we just add a few labels to the OperationalConfiguration and get the same thing?

Is someone working on defining an Application Metadata scope? Currently, only Network and Health scopes are defined.

Here are all of the usages: ``` 2.overview_and_terminology.md 21:> A cloud native application is a collection of interrelated, but discrete _components_ (services, tasks, workers) that, when coupled with configuration and...

Pursuing Option 2 would require switching the entire platform from asynchronous to synchronous. In this case, instead of using state resolution loops, we will have to do incremental deployments and...

Interesting idea. My intention with implementing ComponentInstance in Kubernetes was not to introduce a new spec item, but just to give an accounting object to Kubernetes users. What implications would...

Anyone working on a non-Kubernetes implementation, can you weigh in on this?

I think when we discussed this months ago, the feeling was that infrastructure operators are typically the ones who decide which storage backs which things. Again, I'm not sure how...

Yeah, adding a mode on top of it that would cover the base cases -- especially for standard CI/CD -- would be good.