Some remarks on Diophantine equations and Diophantine approximation (Q763582)

Let \(S\) be a finite set of places containing the Archimedean places of a number field \(K\) and denote by \(O_S\) the ring of \(S\)-integers. Call a triple \((x,y,\varepsilon),(x',y',\varepsilon')\in O_S^2\times O_S^*\) equivalent if there exist \(\eta\in O_S^*\) and a rational integer \(m\) such that \(x'=\eta x, y'=\eta y\) and \(\varepsilon'=\eta^m\varepsilon\). In the paper under review the authors consider the following four Diophantine assertions: 1.) Let \(F\) be any binary homogeneous form over \(K\) such that the polynomial \(F(X,1)\) has three distinct roots in \(K\) and \(k\in K^*\) be arbitrary. Then the Thue-Mahler equation \(F(X,Y)=kE\) has only finitely many equivalence classes of solutions \((x,y,\epsilon)\in O_S^2\times O_S^*\). 2.) The Thue-Mahler equation \(XY(X-Y)=E\) has only finitely many equivalence classes of solutions \((x,y,\epsilon)\in O_S^2\times O_S^*\). 3.) The \(S\)-unit equation \(E_1+E_2=1\) has only finitely many solutions \((\varepsilon_1,\varepsilon_2)\in O_S^*\times O_S^*\). 4.) Any set of \(S\)-integral points of \(P^1(K)\) minus three points is finite. The authors prove by elementary means the equivalence of these assertions. The authors point out that since each of the four assertions is a theorem, where effective results are known, one can deduce from one effective version from one theorem effective versions of the other theorems. The authors also establish similar results for higher dimensional spaces.