-
references
-
goals
- cleanest possible api to reflect data
- most useful (minimal extraneous stuff) data returned for interface-purpose
-
I'm using the term
referent_relationshipa lot because there's a world of interesting things we could do with the concept of relationships apart from individuals. Example:'Newport' --> 'had-the-name-of__by-local-native-americans__in-1750' --> 'X'. So I'm explicitly reminding myself that this relationship project is not that.
- GET
/referent_relationships/list/- (don't think this is useful -- putting it here for mental exercise completeness)
{
'meta': {
'count': 1,234,
'range_start': 0,
'range_end': 100
},
'relationships': [
'https.../referent_relationships/123',
'https.../referent_relationships/124',
etc
]
}
- GET
/referent_relationships/123,124/
{
'meta': {
?
},
'referent_relationships': [
{
'relationship_id': '123',
'subject_referent': 'https.../referent/987',
'object_referent': 'https.../referent/783',
'relationship_string': 'is-the-same-person-as', # always this, yes? Not, eg `Father of`
'confidence': 'high',
'approved': 'yes'
},
{
'relationship_id': '124',
'subject_referent': 'https.../referent/321',
'object_referent': 'https.../referent/956',
'relationship_string': 'is-the-same-person-as',
'confidence': 'medium',
'approved': 'no'
},
]
}
- BIG QUESTION -- are we ONLY having the single ``is-the-same-as` relationship for this referent-relationship project? And letting stuff like 'Father-of' be inferred (for the other is-the-same-as person)>
- Only relevant if we're doing stuff like "Father of" -- type' could be one of: [ 'document', 'inferred_auto', 'inferred_candidate', 'inferred_approved' ]
- Q: do we want to track the user that establishes the relationship, and the user who 'approves' it?
- GET
/referent_relationships/referent/123/
{
'meta': {
'count': 6
?
},
'referent_relationships': [
'https.../referent_relationships/123',
'https.../referent_relationships/124',
etc
]
}
- possible idea (don't think it's a good one) -- break out the referent_relationships by type, so the list of explicitly-entered ones from the document, and then the list of inferred-ones, etc -- with counts reflecting those.