Fable
The great clock was not recast; only a few fine gears were replaced
The bell tower city had a giant bell that could announce the time, fires and tides. The bell body was heavy and the gears complex; the old craftsman spent ten years tuning it until the whole city could hear it accurately.
Later, when the new port opened, the city needed one more short bell tone to remind people of the tide gate. The young craftsman said they might as well recast the entire bell and replace all the gears. The city lord calculated the cost, and it was almost equivalent to building another bell tower.
The first fix was to have the bell ringer memorize the new rhythm. It worked during the day, but at night when a fire alarm and a tidal event occurred simultaneously, the bell ringer was overwhelmed, and the old timekeeping was disrupted as well.
The old craftsman did not touch the clock body. He opened the side case and found several small tooth positions that could slightly affect the rhythm, adding a few thin gears so they intervened only during tidal lock, while normal timekeeping still used the original large structure.
The novice worried that a few small gears were too weak. The old craftsman told him to listen to the night bell: the small gears were not responsible for rebuilding the entire sound, but only applied subtle shifts at key turning points, enough to let the old bell learn a new rhythm.
The first trial run was still incomplete, and the tide-gate bell had a bit of a trailing echo. The old craftsman adjusted the meshing of two thin gears again so that the new rhythm fit the port's needs better, without affecting the fire alarm or the hourly chime.
The city lord finally understood that when facing a huge and expensive legacy system, adapting to a new task does not necessarily require recasting the whole body. Sometimes you only need to add small trainable components at a few low-cost positions.
Later, neighboring cities also came to ask for clock modifications. The old craftsman did not dismantle every clock into parts; instead he first judged whether the old clock's main body was reliable, then decided where the thin gears should be added. If the old clock itself was broken, small gears could not save it; if the task was only to learn a new chime, small components could bring change at very low cost. The city lord also kept the old clock's main blueprint, with the thin gears registered separately. If the tide-gate sound was no longer needed someday, they could be removed without damaging the original clock. Small modifications are useful precisely because they are light, controllable, and fitted to the new task, rather than rewriting all history. Later, when townspeople heard the new chime, they could still recognize that old clock, only now it knew one more thing. This sentence was later written into the newcomers' handbook, reminding everyone to look at the on-site cost first before talking about saving effort. Because of this reminder, when similar difficulties arose later, people did not rush to apply old methods. They learned to ask first which step bears the risk, then decide whether to loosen, tighten, or rearrange the process.