Forking becomes less likely/more costly as a system is widely adopted, and as it is widely decentralized. I agree with you on historical precedent but have more faith in this for a variety of technical and structural reasons. Disclaimer that I’ve known Arthur and Kathleen for a long time (pre-Tezos) and have helped out a little with some infrastructure/security a few times with the project, but that also means I have seen how the sausage is made.

My “article of faith” is that one time someone will get this sufficiently right to facilitate a lot of awesome stuff. Based on what I’ve seen, I think Tezos has a good chance of being “it”. People should definitely do their own research, and it isn’t an all-or-nothing trust thing; trust can be built in a network over time and through use.