I feel like you don't understand the position because there is nothing in what you're saying that implies that you do.

I'm going to play this conversation as it occurred from my perspective to see if you see what I mean.

Your first response is "you're taking an absurdist position, so I'll take the opposite absurdist position to demonstrate the problem. Could we eliminate all racist rules, of course not. Car rules can be racist, but we can't just not have car rules"

I reply "yeah, but we can not have cars. Cars aren't a requirement for society"

You reply "but rules would still apply to those who do the not-car transport"

I reply "yes, but that wouldn't exclude them from society. They would still be able to participate, unlike those kicked out of the hypothetical store"

To which you reply "but the grocery store wouldn't apply to everywhere"

And I retort "no, but if they had any popularity, they would expand in order to deny disadvantaged people groceries at these 'better' stores"

And then your latest reply, which I can't summarize without it becoming a straw man (my failing, not necessarily yours).

This grocery store isn't "people extending basic decency" it's "people not inconveniencing others on threat of permanent removal". One is a social contract extended by and agreed to by others (basic decency) and the other is a threat enforced by the system, in this case the grocery store. You're arguing that systems need rules. I'm arguing that using systems when it could just be standard human interaction is insane. Do you see the disconnect now?

Systems should be built to accommodate humans, not replace human interaction. Jane paying with a checkbook isn't a reason she be barred from a public service. Christ on bikes, man.