Fable
Choosing messengers cannot rely only on who runs fastest
In its early years, White Tower Town had only one street. The mayor chose messengers very simply: candidates ran from the south gate to the north gate, and whoever arrived first was responsible for delivering letters. At that time there were few letters and the road was straight, so this method was sufficient.
Later the town expanded to the hillside and the river port. Letters included urgent letters, bills, secret missives, and prescriptions; roads included stone steps, mudflats, narrow bridges, and night paths. North-south gate sprints could no longer represent all capabilities.
The mayor's naive fix was to lengthen the track. The candidates did indeed get more tired, but it was still impossible to tell who could find the pharmacy on a rainy night, or who could report promptly when a secret letter was lost.
Worse still, some people specifically trained on that long track. They knew where the pits were and where shortcuts could be taken, and their scores grew more and more impressive, yet when it came to actually delivering letters they often got lost.
The old postman suggested first writing down clearly what the town truly needed: punctuality, accuracy, confidentiality, handling surprises, and explaining failures. Then design fixed scenarios for each type of capability.
The new test course included rainy roads, wrong house numbers, duplicate addresses, temporary bridge closures, absent recipients, and broken seals on confidential letters. Each candidate ran the same set of scenarios, and time, errors, remedies, and delivery results were recorded.
Some people come first in long-distance running but give up immediately in front of the wrong doorplate; some are average in speed but verify the address and leave records. The mayor could finally compare which types of tasks different couriers were suited for.
The testing ground was not set in stone either. After the town added a new dock, tides and shipping schedules were added to the evaluation; after scam letters increased, identity verification was added. From then on, White Tower Town no longer decided trust based on a single performance. It used a set of repeatable, comparable tasks close to real work to judge who could take on which kind of responsibility. Later, horse-riding messengers and ferry messengers arrived in town. If they only ran the north and south gates, they could all beat the walkers; but when faced with hillside alleys, closed bridges, and recipients not at home, their advantages immediately changed. The mayor then separated the scores: speed, accuracy, remediation, confidentiality, and cost each got their own column. Messengers no longer competed for a single overall champion, but looked at which type of real errand they could reliably handle. The mayor also ruled that candidates could not know all the routes in advance, only the evaluation rules. This way, what was compared was not memorizing the track, but completing real errands under the same pressure. Later, all new routes first entered the testing ground. The testing ground thus became more and more like a real town.