Hit Song Science?

I completely disagree with the Hit Song Science software. While it is an interesting concept, I do not think that a computer should be the one deciding whether a song is good or bad. Those decisions should be left to the fans. If a song is a no go according to actual people, then don't release it. The whole point of being a musician is seeing whether or not one has what it takes to please a large group of people, not what it takes to please a machine. The software is not the one who will be buying the artists' albums, so why should its opinion be so important? Why should it be the one making the decisions when it comes to releasing a new song? Plus, people have all types of tastes in music, and the software is just making a generalization based on previous songs. Most people like hearing a different song. I do not think a lot of people would be happy if every song they heard was a bit too similar to the last one. What if the software is wrong? The software might give a low rating to what could have been the next biggest hit across all genres of music and give a high rating for a song that people might end up hating. When it comes to likes and dislikes about music, opinions should be left to people and fans with emotions, not wires and microchips.