That’s ok, since I’m internment the current model and creating educational materials for that, the issue I see is not having any real examples of what an AGENT is in the current demographic model.
The second open question is what’s the place of devices in the current models (the things I mentioned above).
We use AGENTs to record all those (autonomous) software entities that perform (internal) tasks or commit data changes. For example we use them for data conversions, imports, exports, migrations, merges or anonymization. These are all executed by (unattended) background processes that have a certain job (script) and a scheduling schema (thus a sort of cron-jobs).
Thanks @sebastian.iancu
Do you use those job agents for the committer or composer? I mean to identify them, and associate to, party proxies?
These are details that I don’t know the answer that quick anymore… I know that it depends on the use-case, as sometimes we need to use a given composer (so as a “on behalf of”), and most of the time the committer is the agent itself - and we use an associated PARTY_IDENTIFIED, or PARTY_SELF if is the patient.