The article states: >"The average post got a score of a few hundred, according to the documents. But in 2019, a Facebook data scientist discovered there was no

The article states: >"The average post got a score of a few hundred, according to the documents. But in 2019, a Facebook data scientist discovered there was no limit to how high the ranking scores could go." Might someone be able to say how the fact that scores could be unlimited was only discovery well after this system was implemented? Wouldn't the upper and lower bounds for scoring be part of the algorithm design?