This means that the binary data can only be decoded correctly if the code reading the data is using the exact same schema as the code that wrote the data. Any mismatch in the schema between the reader and the writer would mean incorrectly decoded data.
