Chris Lauwers

As a rule of thumb, TOSCA uses lists whenever the **order matters**, which is definitely the case for requirements. A type designer may require that an orchestrator fulfill requirements in...

In my implementation, I encode "complex" values (i.e. maps, lists, and complex data with sub-properties) using JSON, but I pass "primitive" values un-encoded. Perhaps we should specify that all values...

If we did this, we would also need grammar for accessing artifacts at the topology template level. The TOSCA language currently includes the following two mechanisms for referencing artifacts: *...

The TC has approved Tal's proposal. https://docs.oasis-open.org/tosca/TOSCA/v2.0/csd05/TOSCA-v2.0-csd05.html#_Toc125468999 needs to be updated to reflect support for tarballs.

I'm not a big fan of the term "unit". As far as I know, it is a term that is not used anywhere else in the industry, and I believe...

In the Ubicity implementation, I don't require that the 'topology_template' must be specified in the "root" file. It can be specified in an imported file, as long as there is...

OK, let's try to decide on Tuesday.

Thanks Mikhail for providing these use cases. You are correct that the TOSCA v1.3 syntax does not allow you to unambiguously define intended requirement mapping behavior. I believe there are...

As far as I know, ETSI NFV only defines policy names without actually expressing any policy semantics. We need to beef up TOSCA support for defining policy semantics (not just...

I have a translator that generates TOSCA profiles from YANG modules. An ‘any’ type would be useful when translating YANG unions. From: Philippe Merle ***@***.***> Sent: Friday, March 11, 2022...