alt Hacker NewsWhitelists have the exact same problem you're objecting to. Not everyone will agree what should or shouldn't be on one.
In practice I don't think it's an issue. What I'm arguing for is the infra to facilitate self categorization and (likely) also a legal requirement limited to only a few specific categories. For example the government might mandate that porn, social media, and user generated content all be accurately tagged and provide legal definitions.
Nothing about what I describe would preclude additional layers of categorization such as (but not limited to) whitelists. In fact it should improve such efforts by providing a standardized method they can use for arbitrarily fine grained categorization that will be compatible with other software out of the box.
Note that my tagging proposition could be applied per network request. So if the service operator wants to it should facilitate filtering out (for example) a comment section without blocking access to the rest of the site.
the point being that instead of there being a kind of commission to create a schema, there are a whole bunch of different whitelists. so if your religion objects to the existence of mangoes, then you can subscribe to a mango-free internet filter.
and instead of burdening the isp the publisher of mango sorbet recipes with ticking off all the right schema boxes, this can all be enforced at the consumer.
all the rest of these approaches kind of assume that there is 'reasonable' and 'unreasonable' content, and that we all mostly agree on the difference. which I think is fundamentally fallacious. do you really think we can agree, as a species, what PG-13 should mean for the entire internet?