possible duplicate of Why is $\sum_{n=1}^{\infty} x^n-x^{n+1} $ not uniformly convergent in $[0,1)$?
I'm trying to understand where exactly I went wrong trying prove this:
"Prove/refute: $\sum_{n=1}^{\infty}(x^{n} - x^{n-1})$ uniromly convergent over $[0,1)$"
So by Cauchy criterion a series is uniformly convergent iff for every $\epsilon>0$ there exist $N$ such thart for every $m>n>N$ and for every $x\in I$ takes place that $|(\sum_{k=n+1}^{m}f_{k}(x)|<\epsilon$
So I did this: Let $0<\epsilon$. There exists $N \in \mathbb{N}$ such that for every $N<n<m$ and for every $x \in [0,1)$:
$|\sum_{k=1}^{m}(x^{k} - x^{k+1}) - \sum_{k=1}^{n}(x^{k} - x^{k+1})|$
$<|\sum_{k=n+1}^{m}x^{k}(1-x)| <_{0\leq x<1} |\sum_{k=n+1}^{m}(1-x)|$
$ < |(m-n)(1-x)| = |m-n+ x(n-m)| $
$<_{0\leq x<1} |m-n+1(n-m)|$
$ = m-n+n-m = 0 < \epsilon$
So on one hand it seems like it is correct while here and also in my study books its not but I couldn't find where exactly I gone wrong.
You cannot possibly prove that $|m-n+x(n-m)|<0$, since no absolute value is negative. Your error lies in the assumption that $|m-n+x(n-m)|$ is smaller than the value that it takes when $x=1$.