Not every pseudoalgebra is equivalent to a strict one (Q664768)

The context of this paper is that of \textit{R. Blackwell, G. M. Kelly} and \textit{A. J. Power} [J. Pure Appl. Algebra 59, No. 1, 1--41 (1989; Zbl 0675.18006)]. For a 2-monad \(T\) on a 2-category \(\mathcal{K}\), it is often the case that the inclusion of the 2-category \(\mathcal{K}^T = T\text{-}\mathrm{Alg}_s\) of (strict) Eilenberg-Moore \(T\)-algebras into the 2-category \(\mathrm{Ps}\text{-}T\text{-}\mathrm{Alg}\) of pseudo-\(T\)-algebras and pseudo-\(T\)-morphisms has a left adjoint. Saying that the unit of this adjunction is an equivalence is a kind of coherence theorem; it implies that every pseudoalgebra is equivalent to a strict one. One might suspect this to be the case in the favourable circumstance where \(\mathcal{K}\) is a locally finitely presentable 2-category and the endo-2-functor \(T\) has finite rank. The author provides a natural counterexample motivated by higher category theory. The paper prepares the ground for this with some basics of 2-monad theory and enriched category theory.