It's no secret that chess at the highest level is one giant research project. The pros enjoy it, if they didn't there's just no way they'd spend so much time/energy.
As for us amateur, I see enough novelties on move 3-5 that I don't know how to deal with -- certainly still complex enough for me.
It's not a bad idea to try and "infuse some novelties" by changing it up, but this argument really only works at the pro level.
I see the other thread about 960 got locked, as the participants descended to trading insults.
Now I'm watching Guseinov vs. Nakamura, last round of the WTC. They're 18 moves into their game - my database shows 75 games with the position they are in, and tells me that the next move will be 19. f4.
Naka spends 9:58 before playing the move the database told me he would play (mind you, I'm not using a computer, it's the database culled from previous high-ratings games to the identical position.)
I think there's absolutely room for a variation that infuses chess with novelties again. Naka has spent 43 minutes so far. The database also correctly predicted the reply: 19. ... Rad8.
Now - if this was 960, those moves would have created completely new and challenging positions. And because of the random nature of the starting positions, it would be quite a while before you'd find two similar games.
Just sayin'!