Michael Voříšek

Hi @georgehristov, the ship sailed already too much to be able to resolve the conflicts easily. Can you please rebase/redo this PR on top of the current develop branch or...

> huge BC break wanted, in refs, we clone before and we can fix easily > does not solve issue with cloning Model (and scope) please describe > as it...

please describe > does not solve issue with cloning Model (and scope) I am not strictly againts a flag in a Condition, but only like `allowSystemIfAlwaysTrue` and default to false...

currently, we assume and change behaviour of Field implicitly please reread the whole discussion - I would remove it completely, but if this makes sense for refs and possibly some...

Unrequested mutation of a model is forbidden by design. Definitive fields should be resolved at runtime - they must not be editable (`Field::isEditable()`) and the default field value must be...

> Provides possibility to filter fields by Closure I do not get this - we should get all and filter then, what is the advantage of your approach?

> > I have another idea: > Presently filters are in first argument as array and considered OR. I like minimalistic APIs. We extend Model more and more for BC......

you are right, this is actually hide field (because of system) with nested conditons like (a > 5 or (a = 5)) (the code will trigger becauseof `=` - but...

one possibility, but this can imply default arguments for API etc., which is not fully correct (too implicit) I think we should drop it - what code relies on it?

please post PR to discuss more concrete - save() works on either loaded row/ID or a new one, right? so if this is only a ref issue, we should be...