I think the biggest problem with polemic blogrants like this are that the author has encountered their own personal bankruptcy with ORMs and proceeds to indict

I think the biggest problem with polemic blogrants like this are that the author has encountered their own personal bankruptcy with ORMs and proceeds to indict the entire enterprise of mapping an object-oriented language environment to a relational object environment. Suggesting that ORMs are awful and no one should use them is effectively premature optimization, and we all know the consequences of that. It is also actively ignoring the history of ORM development. Django's ORM, I know for instance, was developed at a newspaper so that you could go from "feature writer has giant data set and interesting idea about how to visualize it" to "production interactive newspaper article up and allowing people to see their community from different vantage point" within the scope of a day or two. And it worked. And it continues to work. Ignoring the functional aspects of the world or proscribing premature optimization with your opinions is effectively soapbox ranting. You can do it, but the solutions and equity being built on the backs of shitty SQL generated by ORMs are going to just walk right on by.