This was discussed on the SNOMED on FHIR call 2019-06-04. The description of the 'reverse' parameter appears partially incorrect, as "This parameter reverses the meaning of the source and target parameters" doesn't seem to be a true statement. When 'reverse = true', the meaning of 'source' and 'target' actually remains the same, but instead the meaning of the 'code' parameter is changed from "The code that is to be translated" to "The code that is translated to". The documentation for 'reverse' should therefore be revised to "if this is true, then the operation should return all the codes that might be mapped to the specified code (this changes the meaning of the code parameter to "the code that is translated to")". The documentation for 'code' should also be revised accordingly, as "The code that is to be translated (or, if reverse = true, the code to be translated to)."

The phrase "... used when the concept (system/code pair) was chosen" in regard to the source value set also seems incorrect or misleading - instead, that sentence probably should be revised to "Identifies the value set that contains the concept(s) to be translated."

The documentation for 'target' may not require any changes, but the documentation for all of the parameters needs to be consistent.