The web of today looks like it does mostly due to it's reliance on open source and 25+ years of public HTTP standards. And some sort of auto synchronizing web d

The web of today looks like it does mostly due to it's reliance on open source and 25+ years of public HTTP standards. And some sort of auto synchronizing web database of this magnitude is probably out of the reach (in time, complexity management, reliance, etc) for the normal open source project. What comes to mind would be something more proprietary like blazor by microsoft. I feel like viewing a web page as nothing more than a "database skin" covers 95% of use cases and obviously SQL is a large part of that. Not sure what your struggle is with tree structures in SQL, that seems like the easiest case. More complicated, as you mention, is multiple M2M rels. Denormalizing these into database views or temp tables can go far, as well as materialized views. Outside of that, populating denormalized dimensional tables via triggers (and other means) goes far as well. This is just kind of how things end up being as data grows and analytics come into play. A good DBA can do wonders, especially as SQL becomes more of a black art due in part to so many software stacks trying to hide so much from the DB layer. So with all that said, I just don't see how trying to do away with so much on the backend and shove it into the browser will in any way "simplify" things.