We are getting deep into safety design factors here. So I appreciate the conversation. No one else is gonna read this. So I'll expand a little. Because you are hitting a good point.

What you're talking about is not another independent point of failure in the scope of "keeping the rider alive". The "scope" is within the design parameters of a locking mechanism.

In my example of power supplies; the redundancy is around the failure of a single power supply. So the scope is "make sure power supplies operate with 'm' levels" of redundancy.

A single portion of the system can have redundancy. And that's what you're talking about in your comment. But, to continue with my comparison on power supplies, the power supply arrays has no ability to ensure it's source power isn't cut. The redundancy in the power supplies is useless if the power source (central power, substation, electrical grid) is cut. It's a single point of failure still.

So, "zooming out" to a more macro level we would want to ensure redundancy there. Backup power generators being the obvious answer.

Do you get my point? There can be redundancy in the single restraint mechanism (just like redundancy in my power supply array) but the "seat belt" is similar to the "power generator". It's another separate level of redundancy.

I hope that makes sense. I zoomed out my analogy from its original intention based on your interpretation of it.

Either way. I like these conversation. When talking about design and safety there is a "scope" that is important to understand. That we assume the same safety requirements are met at the point of which we have no control over or are not considering in that design. It can be outside of the "scope" that the design is meant to control. So, reading a safety standard on a specific locking mechanism is within that "scope" but assumes further safety and redundancy beyond its given scope. Just as we assume a hospital will have backup power generators when designing a single power supply array. A safety standard for a locking mechanism will have redundancy built into it; but assumes, outside of its design scope, that there is further redundancy in "keeping the rider in the seat". Because it's entire goal is "keeping the rider in the seat". The locking mechanism redundancy is tighter, factored around if X part of the mechanism fails then Y part ensures the rider is contained.

Hope that makes sense. All systems will have a single point of failure. The point is to understand the scope of which that system is meant to operate in and add redundancy to it. If we "zoom out" enough all roller coasters safety systems will fail if the ride is hit by an astroid from outer space. The point is in understanding the scope of the system and adding redundancy in multiple independent mechanisms. The scope being "multiple independent systems of restraint". Which is what the seatbelt adds.