Yeah, the stated intention with the Eureka boosts is to ensure that the exact opposite happens - that the "optimal" path really depends on your location, your map, your strategy.
For reference - 3 spearmen in Civ 5 was a lot. Each spearman costs 56 hammers, so 3 spearmen cost 168 hammers, and the three of them would have a maintenance cost of 3gpt. For comparison, the Great Library costs 185 hammers. So if I go for building 3 spearmen, it's costing me basically an entire ancient Wonder worth of production! (A little less production, but has ongoing maintenance cost). Based on these numbers, unless I actually planned to use the three spearmen to wage a war soon or needed them to clear out barb camps, it would almost certainly be a waste of time to build them early just to get half progress on an ancient-era tech.
Obviously, these numbers aren't going to stay the same in Civ 6, but overall the ratios might hold. If that's the case, it looks like the boosts are not going to be worth going too far out of your way for.
Building cities by mountains and coast is also a reasonable example, because it's actually really constraining. My initial city is probably going basically where my starting location was... and it's really important for your second and third city sites to be good too... is it going to be worth it to specifically look for a city site by a mountain for the research boost? Probably not as important as placing that city by a river or by luxuries or strategic resources and amenities.
An individual boost isn't all that huge. It's 50% of one technology. That seems like it's not going to be big enough to do something like determine where you found your second city, it's not something to devote even half of the cost of a wonder to. (Unless you have a specific strategic reason to want THAT tech really fast, or unless you really are willing to devote a huge chunk of your empire's production to speed up your science, at the cost of developing whatever your actual strategic focus is. )
Some interview with the designer did say that the "single optimal strategy" issue was part of why they put in the Eureka moments in the first place, to give a player lots of things to react dynamically to and make sure a player *can't* just decide, at the start of the game, what path they want to take without knowing where they'll find themselves. Obviously, we don't know whether they've succeeded in their design goal until we try the game and see for ourselves, but they are clearly worried about your concern as well and are trying to address it.