How can we relate the two informal definitions of the limit with each other?

Question

Informal definition $1$:

This is the way in which I studied it, and it makes sense to me.

As $x$ gets closer and closer to $a$, $f(x)$ gets closer and closer to $l$.

---

Informal definition $2$:

$f(x)$ can get arbitrarily close to $l$, by taking $x$ sufficiently close to $a$.

---

I am not able to relate these two informal definitions. How do we relate these definitions?

Answer 1

The second informal definition is more precise. Here's the problem with the first one: consider $f(x) = x^2$, and take $x \rightarrow 0$. Now I claim that the limit is $-1$:

As $x$ gets closer and closer to $0$, $f(x)$ gets closer and closer to $-1$.

This statement is true, but the actual limit is certainly not $-1$.

The problem with the first definition is that it's so informal that it's missing the critical part of the concept, which is that I should be able to make it as close as I want to $-1$ by choosing values of $x$ close to $0$ (which I can't in this case). That precision is what the second definition supplies.

Answer 2

The first informal definition suggests motion: as $x$ somehow "approaches" or "gets closer to" $a$, the value $f(x)$ approaches $L$.

The second informal definition is actually the formal definition, written in English. Mathematicians designed it to capture the idea of "approaches" without the need for vague terms like "closer and closer to".

The English phrase "arbitrarily close to $L$" is written more formally as "closer than $\epsilon$ for any particular $\epsilon$ you may have in mind." The "arbitrary" says "for any possible $\epsilon$."

The "sufficiently close to $a$" is written more formally as "you can find a $\delta$ (which will usually depend on $\epsilon$) that will guarantee $f(x)$ within $\epsilon$ of $L$ as long as $x$ is within $\delta$ of $a$." The $\delta$ is sufficient (i.e. good enough) to guarantee the $\epsilon$ closeness of $f(x)$ to $L$.

Answer 3

Both sentence render the part

$$|x-x_0|<\delta\implies |f(x)-L|<\epsilon$$ of the formal definition.

But the definition 1 is lacking an essential ingredient: $\epsilon$ can be as small as we want. Indeed, we could have smaller and smaller $\delta$ corresponding to smaller and smaller but bounded below $\epsilon$.