Scoping dogs many game developers. It's not solely a ego or leadership thing, indies do it to themselves too. The problem is that game scope blows up much fast

Scoping dogs many game developers. It's not solely a ego or leadership thing, indies do it to themselves too. The problem is that game scope blows up much faster than you expect because it's disproportionately based on how many contradictions appear in the initial design. That is, if you introduce overlapping systems like "can traverse spaceship interior and exterior on foot", "can open and close doors and airlocks", and "other players respond to physics", every way in which one system has been hacked into being (which is most of them, in games) will carry over to hacks in the others, with correspondingly more time spent on assets, testing, bugfixes, etc. Repeat that across an "everything" game and you automatically get an impossible project that could only be achieved with a much more intensive simulation everywhere(which would exceed any plausible computation budget). The design has to put its foot down somewhere and say "we aren't going there" to resolve whatever contradictions it can. For example, replacing fully traversable environments with cutscenes and UI, or replacing lengthy animations with iconography. The games that can justify going really deep into their simulation end up being pretty niche, like Dwarf Fortress. But actually making the resolution to settle on a more streamlined focus that would cohere well also breaks the hype of "sheer possibility". As long as you keep adding more, more continues to seem possible. Both creators and fans like staying in that zone. Nobody likes cutting their sunk costs, so they will rush to justify keeping stuff that doesn't work in the game until the project is cancelled due to lack of progress.