And?
The trade-offs between longpolling and websockets are rarely discussed outside of very small cohorts. The tradeoff between wankel-engines and piston-based motors are rarely discussed. The tradeoffs between oak and pearwood are rarely discussed.
That's why our society has division of labor and people becoming experts in their topics.
And I am pretty sure experts on cartography have discussed the different merits and tradeoffs and converged on Mercator for a reason. I don't know what that reason is, and I don't care. It works, I can use it, and as long as no one can point out a tangible problem with it, I see no reason to worry about it. Same as the cartographer doesn't need to know or care why I decided that the chatapp he installed uses longpolling to communicate with the backend instead of a websocket. Same as I don't need to know or care why the carpenter who made my desk used oak instead of pear or rosewood.
And seeing countries as smaller or bigger than they are isn't a problem in my eyes. I can reasonably expect an educated person to know that maps are 3D->2D projections and are thus distorted. If I want to know the exact size, I can always look it up. I don't have a more or less favourable view of a country or its people based on the countries apparent size on a map.