I was skimming through a few books on calculus, and came across the statement that:
we cannot write out this limit as $+\infty$ or $-\infty$ like so: $\lim_{x \to 0}(-\frac{1}{x})=-\infty$, only as $\lim_{x \to 0}(\frac{1}{x})=\infty$. The argument provided was that its a matter of agreement.
Although, it's considered perfectly fine to write this equation likeso: $\lim_{x \to 0}(-\frac{1}{x^2})=-\infty$
My question is whether anyone else has written it in opposition to the statement above, and what is meant by "matter of agreement", and why it may be as such.
Or might it just be the authors own biased opinion, for whatever reason?
I personally see nothing wrong with introducing $+\infty \space or -\infty$ given that it helps to recognise further meaning in supporting your rationale for proving a theorem.
Here's an image of the page where it's taken from:

Reference: Fort, T. (1951) Calculus. pg 50.
Your comments are most welcome.
Presumably, the "matter of agreement" means that denoting $f(x)=-\frac{1}{x}$, you have $$ \lim_{x\to 0+}f(x)=-\infty $$ and $$\lim_{x\to 0-}f(x)=+\infty $$ which do not agree.
Alternatively, the definition of "$\lim_{x\to 0}f(x)=-\infty$" says that
But this is impossible in the case when $f(x)=-\frac{1}{x}$.