Seeing the webapp that @ilgooz built, I was thinking that we can use it in to actually create application (the workflows) directly from the ui.
This will not be permanent running app but it allows us for now to experiment without taking strong decisions for the core. When we are confident that the data structure that we have is good enough then we can extract this and include it in the core.
Also, we discussed with @Nicolas and we end up to the conclusion that it would make sense that the UI only use (system)services and only access to the
executeTask. All the other APIs can be used as (system)services. Of course for now it’s ok to use existing APIs because we don’t have everything as a system service.
Also I’ve been working before on a prototype of UI if this can help for inspiration