Someone asked me whether staking + shared security is really good or not... I usually consider "yield stacking" and "risk stacking" together. To be honest, don’t just focus on the small extra gains. If you give multiple services endorsement for the same principal at the same time, you might get hit all at once if something goes wrong: penalties and confiscations, longer unlock periods, liquidity being locked up, and in the end, liquidation can become very ugly all at once. Recently, before and after the upgrade of that mainstream public chain, people are guessing whether projects will migrate. I actually care more about: will parameters be changed temporarily, will nodes be unstable, will bridges and oracle services get stuck... these are the real costs of "shared security." Anyway, my current approach is pretty simple: only engage in what I can clearly calculate; if I can't figure it out, I just pretend I didn't see it.