I agree a guide should try and be as all encompassing as possible but I think it's also important to have problems to work on where the solutions are not available.

Let me give an analogy: sat nav's clearly are useful devices. I just put in the coordinates of where I want to go, and I get step-by-step instructions how to get there. After following the route a few times, I don't even need the sat nav anymore - I just remember.But if I want to get to a new place, even one fairly close to the first, this knowledge that I've acquired won't be of much use. Now suppose I navigate without using a sat nav. First I have to study a map and plan the route in advance. As I proceed along the route,things might look different to what I expected, and I might get confused, or take a wrong turn. I might have to look at the map again, or ask for help. But the knowledge that I gain at the end of this process can be very valuable. I don't just know how to get to a particular place, but also understand something of the lie of the land, and how that place stands in relation to others.

Another example: as a researcher I try to formulate and/or solve problems for which solutions are not yet available. Sometimes the problems are straightforward, other times not, and I have to work just to come up with an idea of how to start. Sometimes I can find the answer in the literature, but knowing where to look is a challenge, and can involve learning new mathematics and translating my problem into a new language.