In the current Feature Dictionary, there are different entity types that have the same name but different 5-letter codes.
In most cases, this appears to be a literal translation of the UHRB data model where an abstract type and a "real" type have the same name, or where a room type and a construct type have the same name. In a few cases, the indoor item has the same name as something derived from a different standard. In almost all cases, the actual definition text is identical or near-identical for both feature types, resulting in an ambiguity as to which one should be used in any particular context.
Does the CDB 1.2 specification mandate entity names to be unique? The proposed CDB X standard does so. From a conceptual modeling standpoint having entity types with the same name results in ambiguity as mentioned, and at a practical level tools may assume that entity type name is a safe primary key (although most tools probably use the 5-letter code instead).
TODO: Need to determine if NAS defines any of these types as well to deconflict with ongoing NGA work.
Consider: downselecting to only one feature type per name and creating a mapping rule for the other one(s), favoring the GGDM or NAS definition and code where possible.
Examples (will update as I find them):
AL195 Ramp vs. UG017 Ramp
UA004 Furniture vs. UI004 Furniture
UG012 Escalator vs. UO011 Escalator
UI002 Control_Panel vs. UO006 Control_Panel
UG016 Pulpit vs. VO004 Pulpit
In the current Feature Dictionary, there are different entity types that have the same name but different 5-letter codes.
In most cases, this appears to be a literal translation of the UHRB data model where an abstract type and a "real" type have the same name, or where a room type and a construct type have the same name. In a few cases, the indoor item has the same name as something derived from a different standard. In almost all cases, the actual definition text is identical or near-identical for both feature types, resulting in an ambiguity as to which one should be used in any particular context.
Does the CDB 1.2 specification mandate entity names to be unique? The proposed CDB X standard does so. From a conceptual modeling standpoint having entity types with the same name results in ambiguity as mentioned, and at a practical level tools may assume that entity type name is a safe primary key (although most tools probably use the 5-letter code instead).
TODO: Need to determine if NAS defines any of these types as well to deconflict with ongoing NGA work.
Consider: downselecting to only one feature type per name and creating a mapping rule for the other one(s), favoring the GGDM or NAS definition and code where possible.
Examples (will update as I find them): AL195 Ramp vs. UG017 Ramp UA004 Furniture vs. UI004 Furniture UG012 Escalator vs. UO011 Escalator UI002 Control_Panel vs. UO006 Control_Panel UG016 Pulpit vs. VO004 Pulpit