Timeseries collections do not support unique indexes on datafields. If a timeseries measurement includes an `_id` field, it will be treated as a normal data field, and duplicate values can be inserted without error. This differs from standard collections which enforce a unique index on the required `_id` field.