These are chat archives for Sensorica/valuenetwork

Jan 2016
Chris Troutner
Jan 22 2016 03:41
That's a handy guide. This will help give us context for expressing those relationships in the UI.
Is it correct to assume that EconomicAgent is the same as a person?
Bob Haugen
Jan 22 2016 10:40
EconomicAgent can be any of several AgentTypes.
These are user-defined, although we start a new install with Individual, Organization and Network.
In order to have types that we can refer to unambiguously in code, an AgentType also has a partytype, and those are a hard-coded set of tuples: ('individual', ('individual')),
('org', _('organization')),
('network', _('network')),
('team', _('project')),
('community', _('community')),
The first member of a tuple is the internal name, the second member is the external name.
Bob Haugen
Jan 22 2016 10:48
Which also shows some hacky evolution: this is more than you want to know, but we and Sensorica dithered for a long time about what their projects are. They are mostly not like what other people think of as a project: most of them go on forever, and have many goals. Sometimes they do have a project, like the PV characterization, that has defined goals and a defined end. But even then, they can have their own value equation and income distribution, and buy things. They are a context for economic events. So we decided they were EconomicAgents, and renamed 'team' to 'project' for external display.
Anyway, a person is an EconomicAgent of agent_type.party_type == individual.
persons and groups both being Agents is a fairly standard model: for economic ontology, it is an ISO standard.
It makes the model and code for EconomicEvents, which is most of what actually happens, a lot simpler.