There is something to be said for books that are so hard, they engage your "eye of the tiger" response. Often I quit something because I'm frustrated and I thin

There is something to be said for books that are so hard, they engage your "eye of the tiger" response. Often I quit something because I'm frustrated and I think, "if this supposedly gentle thing frustrates me, the real stuff later on is gonna kill me." When you read e.g. baby Rudin, you get a sense that this is the real stuff, and you're getting your ass kicked by the best. However, there is also something to be said for motivation. In some topics, like computability, (much of) what we want out of our theory can be understood in an hour, or at least it feels that way. We can then spend the rest of the time happily quenching our thirst. Not so with analysis. The core problems of analysis seem obscure, or plain incomprehensible, unless you're already on the other side. When it comes to analysis in particular, one approach is to start studying it in the order it was discovered. There's a well-regarded book called Analysis by Its History, but I haven't read it. There's another called Understanding Analysis by Stephen Abbott that I recommend if you want to regroup before attacking baby Rudin again.