I have trouble proving discontinuity of the Dirichlet function, using the $\epsilon-\delta$ approach.
The function is defined as follows:
$$ f(x) = \left\{\begin{array}{l l} 1 &\text{if }x \in \mathbb{Q} \\ 0 & \text{if } x \notin \mathbb{Q}
\end{array} \right. $$
Would it do any good showing the discontinuity at some $x_0$, by bifurcating the problem into two cases, one where $x_0$ is rational and one where it isn't?
The above function isn't continuous anywhere, but let's look at one that is continuous at only one point in its entire domain:
$$ f(x) = \left\{\begin{array}{l l} 0 &\text{if }x \in \mathbb{Q} \\ x & \text{if } x \notin \mathbb{Q}
\end{array} \right. $$
I can see that this function is continuous at 0 alone, but once again, not able to show it rigorously by picking an appropriate $\epsilon$.
How should I go about investigating the continuity of such "weird" functions, using $\epsilon-\delta$ arguments? I'm sure there are many more to add to my troubles, such as Thomae's function, for instance. I'm really more concerned with the approach than with the solution, though it'd be great if someone could help me figure out proper proofs for the above functions, so I can at least get started from where I ended up getting stuck (all the functions look pretty similar in that sense, and knowing how to work with $\epsilon-\delta$ with even one should help me figure out the rest)
Please help me out, I'm pretty new to real analysis! Thanks a lot in advance!
Best Answer
Remember what epsilon-delta continuity is: anytime someone gives an epsilon, you have to respond with a delta that works if you want to prove your function is continuous. Sometimes the strategy can feel a little contrived, but imagining this sort of game can be a real help. When dealing with strange functions, think about the "special features" of your function - what makes it unique? It will take you some time to get used to this, but with practice, you will develop your intuition.
For the first problem, there's no need to use caserwork: it is possible to pick an epsilon so that there is no valid response of delta. Can you think of what it is?
Hint:
Bigger hint:
For the second problem, think again about what you need to do. If you're at zero and someone gives you an epsilon, do you see a choice of delta so that $f(x)$ will be less than $\epsilon$ for all $x$ with $|x|<\delta$?
Big hint: