She's right that if all you need to do is glue stuff together, then she doesn't need to study it. Trivial problems are trivially solved. By very low-paid people

She's right that if all you need to do is glue stuff together, then she doesn't need to study it. Trivial problems are trivially solved. By very low-paid people. Who are very bored. But what I've noticed is that there are only so many non-trivial problems, and the formal pieces are the little pieces of distilled wisdom, the 'lego bricks' from which you build anything beyond trivial software. You might find studying graph analysis algorithms, say, to seem utterly academic, but there will be a time when what you're asked to do is to analyse a graph, and you'll be glad that your 'mental index' includes that entry. I suppose the thing is this; if you've studied enough that ten years down the line, someone can ask you to solve a problem and you can say 'ah, yes, this requires an FSA' or 'I can do it with A*' or 'that's easy with a red-black tree', then you know enough.