colmsnowplow

I need to think it over, as I do also see the merit in the suggestion, but I'm unsure if my concerns on it are legit. However, to play devil's...

Thinking further on it - plugging in custom modules means that we can't know in advance whether or not a custom module exists, nor can we know what the module...

> Would love to dig into an example of where that would be the case! Hard to explain simply, but I'll try. The new web model structure aims to have...

I am 100% on board, by the way, with the idea of removing the complexity of playbooks (and potentially removing the need for a user to explicitly define the playbook),...

Yeah I like this direction. Relevant to the idea, by the way, is that evolving this kind of structure involves propagating playbooks with how SQL-runner currently works - so there...

Noting this here in case I forget. If we pursue the ideas outlined here, it would be nice to be able to both set global vars (for an entire DAG...

> I'd prefer to do that though with an explicit test for table existing rather than catching an exception. It's more deterministic. Yeah for sure I agree. > Agree this...

Interesting. The missing piece of the puzzle that I need to test then is whether this can handle non-existent resources (or checks for their existence).

I like this. A good example of it done well in [great_expectations](https://docs.greatexpectations.io/en/0.9.8/reference/data_context_reference.html)

I hit this problem too. When I navigate to https://releases.hashicorp.com/terraform/ it looks like the issue is with the path - the directories are all `terraform_X.X.X`, rather than just the version....