Gamblers fallacy helps us out here. If winning is more memorable than losing everyone can win sometimes/in some activities and everyone is happy. Also, I wouldn't play a sport in which I always won.

I find these particular questions quite hard to think about, so I'll just mention these few thoughts: * There is a huge difference between wanting to win, and wanting others to lose. Not everyone will be on the same wavelength, but if they're mostly on the first wavelength, it creates an atmosphere of friendly competition / self-betterment, whereas wanting others to lose looks like academia (bitter competition / self-aggrandizement). In the former, you can lose thoroughly and still get satisfaction out of your participation, so I don't hesitate to say that updating in this direction promotes a better social environment for every individual. Perhaps this somewhat answers your third question. * There's almost certainly value in limiting competition size. Losing with a thousand others placing ahead of you is much less motivating than losing with 50 others placing ahead of you. (it's not clear to me what exactly you meant by 'games' here -- the most general sense?). So if your * Many games can also be played with a focus on beating yourself rather than your competitors. Having the mental resolve to do this consistently is relatively rare, but AFAICS this is a strict win (both in satisfaction/motivation levels and quality of results) over merely beating your competitors. Updating in this direction should make you more friendly in competition, more effective, and less vulnerable to temporary setbacks. And also more able to continue improving even if you are ranked at the top. [...] I think if you look at the wild variety of Linux distributions, that effectively answers these questions, assuming you believe that open-sourcing this stuff will be mandatory (I think it must be, in order to avoid social chaos and oppression, but I don't know if it will be). Perfection is highly subjective/contextual, and even transhumanists have limited resources to allocate. There's also a pretty strong argument to be made that once we can 'reallocate' resources like intelligence, phys