5 great vs 1,000 average, it depends on the technology and the application they are working on, and the stage of the product.
First, the logic in the original article is not valid. Sports and software engineering are basically incomparable. What they share as team activities don't justify the reasoning in the article. A simple counter example. In professional soccer games (like in the top leagues in Italy, Spain, Egnland, etc.), a second-tier team of ll players could quite possibly beat a top-tier team of only 9 players (the other two players are kicked out for whatever reasons). In software engineering, it is quite safe to say that 5 great engineers can outperform 30 average engineers.
Back to my original point, it depends on the technology and the application they are working on. If the technology and the application require high creativity and innovation, 1000 average engineers cann't lead it long. Most probably, it will lead to a mess that has to be cleaned up and rewritten later. On the other hand, if the application is already well designed and well-established, and the primary jobs are to add new features within the architecture, 1000 average engineers may deliver more than 5 great ones.
In reality, a good combination of both types might be the best scenario.
First, the logic in the original article is not valid. Sports and software engineering are basically incomparable. What they share as team activities don't justify the reasoning in the article. A simple counter example. In professional soccer games (like in the top leagues in Italy, Spain, Egnland, etc.), a second-tier team of ll players could quite possibly beat a top-tier team of only 9 players (the other two players are kicked out for whatever reasons). In software engineering, it is quite safe to say that 5 great engineers can outperform 30 average engineers.
Back to my original point, it depends on the technology and the application they are working on. If the technology and the application require high creativity and innovation, 1000 average engineers cann't lead it long. Most probably, it will lead to a mess that has to be cleaned up and rewritten later. On the other hand, if the application is already well designed and well-established, and the primary jobs are to add new features within the architecture, 1000 average engineers may deliver more than 5 great ones.
In reality, a good combination of both types might be the best scenario.