A note on \(D\)-spaces (Q5920619)

\textit{E. K. van Douwen} [Pac. J. Math. 81, 371-377 (1979; Zbl 0409.54011)] defined a space \(X\) to be a \textit{\(D\)-space} if for every map \(N:X\to 2^X\) such that \(x\in\mathrm{Int}\,N(x)\) for each \(x\in X\), there exists a closed discrete subset \(D\) of \(X\) such that \(X=\bigcup\{N(x):x\in D\}\). In this paper, the author introduces notions of nearly good relations and \(N\)-sticky modulo a relation as tools for proving that spaces are \(D\)-spaces. As a corollary to general results about such relations, he proves that the space \(C_p(X)\) of all real-valued continuous functions on \(X\) with the pointwise convergence topology of is hereditarily a \(D\)-space whenever \(X\) is a Lindelöf \(\Sigma\)-space. This answers a question of M. Matveev and improves a result of \textit{R. Z. Buzyakova} [Proc. Am. Math. Soc. 132, 3433--3439 (2004; Zbl 1064.54029)], who proved the same result for \(X\) compact. The author also proves that if a space \(X\) is the union of finitely many \(D\)-spaces, and has countable extent, then \(X\) is linearly Lindelöf. It follows that if \(X\) is in addition countably compact, then \(X\) must be compact. He also proves that Corson compact spaces (i.e., a compact space which can be embedded into a \(\Sigma\)-product of real lines) are hereditarily \(D\)-spaces. These last two results answer questions of A. V. Arhangel'skii. Finally, the author solves a problem of E. K. van Douwen by proving that a perfectly normal, collectionwise normal space constructed by \textit{R. Pol} [Fundam. Math. 97, 31--42 (1977; Zbl 0374.54023)] is a \(D\)-space.