The proof of Fermat's Last Theorem (Q1367442)

Even after the proof of Fermat's last theorem by \textit{A. Wiles} [Math. Ann., II. Ser. 141, 443-551 (1995; Zbl 0823.11029)], there may be the (little) hope to find an elementary proof, using only the techniques Fermat presumptively had known. But the approach of the author is poor. He constructs an integer polynomial \(f(S,Y)\) in unknowns \(S\) and \(Y\). Setting \(x=sy\), where \(x,y,z\) are integers with \(x^n+y^n=z^n\) he believes that \(f(s,Y)\) remains an integer polynomial in \(Y\). So we have the mistake that if \(f(s,y)=c\) equals a nonintegral rational \(c\) than \(y\) must be nonintegral rational. So we would have a proof of Fermat's last theorem for \(y\mid x\) (which is not hard). But the reviewer thinks the author does not even show this, because he never proves that his \(c\) is nonintegral rational, but wants to enforce this by changing some constants which change \(c\) but which also change \(f\).