Re-staking / shared security, put simply, is taking the same collateral and using it to “endorse” more places. The returns may look like they can be stacked, but the risks stack too—it's just that many people don't like to calculate them. Add a slashing condition to the patch, tweak a penalty window, and on-chain it looks like just a few lines of parameters; but once it hits your position, it can feel like, “I didn’t do anything—why am I being penalized?”

Right now, when I look at a project, I first read the update log and the security announcements, especially those points that can set off a chain reaction—external dependencies, oracles, cross-chain integrations. If I can’t make sense of them, I’d rather take less.

On a related note, I also see the community arguing about the privacy coin / mixer compliance boundaries—it’s quite similar: everyone wants “functionality” and “returns,” but when something really goes wrong, how responsibility is divided is often never clearly explained… Anyway, read the code first, and then talk about conviction.