Namioka spaces, \textit{GO}-spaces and an \(o\)-game (Q683958)

The class of \textit{generalized ordered spaces}, or GO-spaces, is the class of all topological spaces which can be embedded in some LOTS -- where LOTS is the class of linearly ordered sets endowed with the usual open-interval topology. Equivalently, \(X\) is a GO-space if there exist a linear order \(\leqslant\) on \(X\) and a base \(\mathcal{B}\) of the topology of \(X\) such that every \(B \in \mathcal{B}\) is an interval in \((X,\leqslant)\). Given topological spaces \(X,Y\) and \(f: X \times Y \to \mathbb{R}\), \(f\) is said to be \textit{separately continuous} if for all \(x_0 \in X\) and \(y_0 \in Y\) the functions \(x \mapsto f(x,y_0)\) and \(y \mapsto f(x_0,y)\) are continuous on \(X\) and \(Y\), respectively. There are several well-known examples of separately continuous functions \(f:\mathbb{R}^2 \to \mathbb{R}\) which are not continuous. A topological (Baire) space \(X\) is said to be a \textit{Namioka space} if for every compact space \(Y\) and every separately continuous function \(f: X \times Y \to \mathbb{R}\) there is a dense \(G_\delta\) set \(A \subseteq X\) such that \(f\) is continuous at every point of \(A \times Y\). \textit{J. Saint-Raymond} in [Proc. Am. Math. Soc. 87, 499--504 (1983; Zbl 0511.54007)], introduced the following topological game (the so-called \textit{\(\sigma\)-game}), which is played by two players, ONE and TWO, on a fixed topological space \((X,\tau)\). The moves of ONE are non-empty open subsets of \(X\), and the moves of TWO are elements of \(X \times \tau\). More precisely, the first move of player ONE is a non-empty open set \(U_0 \subseteq X\). The first move of player TWO is a pair \((x_0,V_0) \in X \times \tau\) such that \(x_0 \in V_0 \subseteq U_0\). Then ONE plays a non-empty open set \(U_1 \subseteq V_0\), and TWO plays \((x_1, V_1)\) such that \(x_1 \in V_1 \subseteq U_1\), and so forth. Player TWO wins the game \((U_0, (x_0,V_0), U_1, (x_1,V_1), U_2, (x_2,V_2), \ldots)\) if the sequence \(\langle x_n: n < \omega \rangle\) has at least a cluster point in \(\bigcap\limits_{n < \omega} V_n\); otherwise, player ONE wins. Saint-Raymond [op. cit.], has shown that if ONE has no winning strategy on the \(\sigma\)-game on \(X\), then \(X\) is a Namioka space. The \(\sigma\)-game is a modification of a game previously introduced by \textit{J. P. R. Christensen} [ibid. 82, 455--461 (1981; Zbl 0472.54007)], and both Saint-Raymond's and Christensen's games are understood as modifications of the classical Choquet game. In the paper under review, the author introduces a new variation of the Choquet game, which he calls the \textit{\(o\)-game}. It is shown in the paper that this game coincides with the \(\sigma\)-game in the class of GO-spaces. The author also investigates the class of GO-spaces which simultaneously \((i)\) satisfy the Namioka condition; and \((ii)\) can be represented as the union of \(\aleph_1\) nowhere dense sets, and shows that this class may be characterized using the \(o\)-game.