Let $C([0,1])$ be the set of all continuous real-valued function on $[0,1]$, let $d(f,g)=\int_0^1 |f(t) - g(t)|dx$.
We note that the exercise 14 needs exercise 13, which gives us a sequence $(f_n)_{n\in\mathbb{N}}$ of the form $$ f_n(t)= \left\{ \begin{array}{lcc} n & if & 0 \leq t\leq n^{-2}, \\ \\ t^{-1/2} & if & n^{-2} \leq t\leq 1. \end{array} \right.$$ which is a Cauchy sequence on $C([0,1])$. Show that this Cauchy sequence does not converge.
Here we are using the following ideas:
Suppose that this sequence is convergent, i.e, there exists $f\in C([0,1])$ such that $f_n$ converges to $f$. Then, we have $$\lim_{n\to+\infty}\int_0^\alpha |f_n(t)-f(t)|dt =0\quad\text{and}\quad \lim_{n\to+\infty}\int_\alpha^1 |f_n(t)-f(t)|dt =0$$ for all $0<\alpha<1$ and $\frac{1}{n^2}<\alpha$.
But let us observe that $$\lim_{n\to+\infty}\int_0^\alpha |f_n(t)-\infty|dt$$ does not exists and $$\lim_{n\to+\infty}\int_\alpha^1 |f_n(t)-t^{-1/2}|dt =0.$$
Therefore, there not exists $f\in C[0,1]$ such that the sequence $f_n$ converges.
(note that you are asking no question, so I'm just guessing what you want here)
As written, the argument makes no sense. You don't say what $\alpha$ is, and you seem to assume that the continuous function $f$ assumes the non-existent value $\infty$ on $[0,\alpha]$.
Instead, what you need to notice is that $f_n\to f$, where $f(t)=t^{-1/2}$, that limits in a metric space are unique, and that no continuous function on $[0,1]$ can agree with $f$ on $(0,1)$.