I don't want to argue the point (but I also am not dismissing your points, the position I put forward is definitely not unassailable), but I think there's an opportunity to make my original point better here, which is; sometimes we get trapped in the logic of our own systems and fail to think outside the box. Is it monoculture or low-N culture really required? Or is it a local optimum we lack the imagination to see beyond?
What got you here might not get you there. You can go really far with a monolithic web app running on top of a relational database. But if you scale far enough, you'll need to pull some pieces out and hook them up to databases with relaxed constraints.
There are good engineering reasons for us to do things the way we do them, and maybe it was the only feasible way for us to get to this point. But presumably if we continue to grow, we will enter a different phase with a different set of tradeoffs. That phase will probably involve exerting less control, it will probably also involve worse unit economics, but may also scale further with fewer externalities.
What got you here might not get you there. You can go really far with a monolithic web app running on top of a relational database. But if you scale far enough, you'll need to pull some pieces out and hook them up to databases with relaxed constraints.
There are good engineering reasons for us to do things the way we do them, and maybe it was the only feasible way for us to get to this point. But presumably if we continue to grow, we will enter a different phase with a different set of tradeoffs. That phase will probably involve exerting less control, it will probably also involve worse unit economics, but may also scale further with fewer externalities.