The repository contains the aspect models based on SAMM (Semantic Aspect Meta Model) for the Tractus-X project for Catena-X.
Currently, we assume the usage of the version 2.9.7 of the SAMM-CLI and version 5.6.0 of the Aspect Model Editor.
The models can locally be processed with the SAMM CLI, which is documented here. It allows you to generate different artifacts (diagrams, example payload, java class files) out of it.
For convenience you can also look into the gen folder of each model, which already contains often used artifacts generated from the model.
Each model has its life cycle and can thus have a different status. This information always corresponds to a specific version of the model. To indicate the state of the model version in Git, one creates a file with the name metadata.json and places it in the same directory as the corresponding model file. It is also possible to load the models into the Semantic Hub which has a corresponding state management. An example metadata.json looks like this:
{ "status" : "deprecate"}
The following table lists the possible values for status and what they mean:
| status | status in Semantic Hub | description |
|---|---|---|
| draft | DRAFT | This version of the model is under development and can change at any time. |
| release | RELEASED | The version of the model is considered as stable and any modifications to the model trigger a new release and subsequentially a new version. |
| standardize | STANDARDIZED | The version of the model is the basis for an official Catena-x standard, published here. |
| deprecate | DEPRECATED | The version of the model has reached its end-of-life and should not be used anymore because it will be deleted later. |
The metadata.json is only relevant for model files on the branch main. All other branches are development branches, and the models from these branches or forks are therefore implicitly in status "DRAFT".
We have a governance process for the joint development of new and updated models which we describe in more details under this link.
The governance process is triggered by a domain expert requesting a new model or model update. A modeling team then reviews the request to identify whether it is of interest for Catena-X and Tractus-X and whether there is not already another model which can solve the raised issue (indicating label MS1-Approved).
Once the request is accepted, a modeling expert and the requesting domain expert create a solution. The modeling expert then evaluates whether the new or updated model follows the modeling guidelines in Tractus-X (indicating label MS2-Approved). In the last step, the requesting use case and the modeling team approve that the resulting model fulfills the initial requirement and can be adopted (indicating label MS3-Approved). As a result, the content of the new model version cannot change, and adopters from use cases are safe to use the model.
We do this process for each version of a model. So there can be multiple versions of the model with different content in different phases of the model life-cycle.