On absolute linear Harbourne constants (Q1747935)

In the paper under review, the authors study the so-called absolute Harbourne constants of line arrangements in projective planes defined over arbitrary fields. Let \(\mathcal{L} = \{\ell_{1},\dots,\ell_{d}\} \subset \mathbb{P}^{2}_{\mathbb{F}}\) be an arrangement of \(d\geq 3\) lines and we denote by \(t_{r}\) the number of \(r\)-fold points of the arrangement, i.e., points where exactly \(r\) lines meet. The linear Harbourne constant of \(\mathcal{L}\) is defined as \[ H(\mathbb{F}, \mathcal{L}) = \frac{d - \sum_{r\geq 2} t_{r}}{s}, \] where \(s\) is equal to the number of singular points of \(\mathcal{L}\). Next, the \(d\)-linear Harbourne constant is defined as \[ H(\mathbb{F}, d) = \min \, H(\mathbb{F},\mathcal{L}), \] where the minimum is taken over all line arrangements of \(d\) lines with \(d\) fixed. Finally, the absolute linear Harbourne constant of \(d\) lines is the minimum \[ H(d) := \min_{\mathbb{F}} \, H(\mathbb{F}, d) \] taken over all fields \(\mathbb{F}\). Let us define the set \(Q = \{q = p^{r}, \, p \, \text{is \, prime}, \, r \in \mathbb{Z}_{> 0} \}\). The main result of the note can be formulated as follows. Main Theorem. For \(d \geq 6\) one has \[ H(d) \geq \frac{1}{2} - \frac{1}{2}\sqrt{4d-3}. \] For \(d = q^{2} + q + 1\) with \(q \in Q\) we have the equality, and in this case \(H(d) = -q\) obtained by the arrangement consisting of all lines in the finite projective plane \(P^{2}_{\mathbb{F}_{q}}\). Moreover, the authors provide the exact values of \(H(d)\)'s for \(d \in \{2, \dots, 31\}\). The authors join these results with the bounded negativity conjecture, but there is no good explanation why such results would be helpful in that context. Also, as a warning, the authors misused the notion of fake projective planes at the end of article (these are by definition minimal complex projective surfaces with \(c_{1}^{2} = 3c_{2}\), and by Kobayashi Hyperbolicity Theorem, these surfaces do not contain any rational curves), and this can lead to possible confusion.