A functional equation of J. A. Baker involving integral means (Q1895208)

At the thirty-second International Symposium on Functional Equations \textit{J. Baker} posed the following two problems [Aequationes Math. 49, 191 (1995)]: 1. Are there continuous, nonaffine functions \(f : \mathbb{R} \to \mathbb{R}\) such that \[ \int^{x + 1}_{x - 1} f(t)dt = 2f (x)\text{ for all }x \in \mathbb{R}?\tag{*} \] 2. Is there an analytic nonaffine \(f : \mathbb{R} \to \mathbb{C}\) such that \((*)\) holds? These problems were solved affirmatively during the Symposium: the first one by \textit{W. Walter} [Aequationes Math. 49, 191-192 (1995)] and the second one by \textit{J. Aczél} and \textit{W. Förg}-\textit{Rob} [Aequationes Math. 49, 195-196 (1995); 198-200 (1995)]. This paper brings details connected with the author's solution of the first problem and, for convenience, a sketch of a proof of the (well known) theorem that the function \(\text{sinh} : \mathbb{C} \to \mathbb{C}\) has a countably infinite number of fixed points. (According to the above mentioned remarks by J. Aczél and W. Förg-Rob, if \(\text{sinh} \lambda = \lambda\) then \(f(x) = e^{\lambda x}\) solves J. Baker's second problem).