Is there a plan to allow customizations/additions to the LDM per workspace in the multi-tenant hierarchical design? We have use cases where we would like to have specialized data models per customer; The underlying schema would be the same but the LDM model that shows up in the data model view would include custom datasets for some customers.

Even now you can achieve this, but only through APIs (or python SDK). You can PUT additional attributes or whole datasets into child workspaces and it should work out of the box. I am not 100% sure that the final model users will see in GUI LDM Modeler app in the child workspace will be correct, but it should be. Please, test it

Thank you I will test that out and let you know

It should work, we cover this use case in our tests, e.g.:

@Jiri Schmid any guidance on what @Jan Soubusta is referring to?

I reviewed the corresponding design document. In the future, we no longer generate these prefixes. In the case of conflict between parent and child IDs, we will apply "parent-win" strategy. We introduce new APIs reporting conflicts. Finally, we provide a new option to specify an object prefix setting on workspace level(optional). Once specified, we will check all incoming object IDs and return user errors if these object IDs are not aligned with the prefix setting.

@Jan Soubusta would it be possible to provide an example to make this clearer?

Generally, in the future object IDs(e.g. fact ID, label ID, metric ID, ...) will never contain the workspace prefix. Both entities and layout APIs can now return object IDs with the prefix, if you work with child workspaces.