[vistakric]

Do I think that it is a good startup model (i.e. something that is likely to get used a lot and/or generate profit)? Yes. Do I think that it is conducive to the goal of acquiring the problem-solving skills to hack something together? Maybe, maybe not.

One of the challenges that a novice developer needs to face and overcome more or less independently is the challenge of making sense of something that looks totally foreign with minimal guidance, and figuring out how this new piece of software fits within their understanding of computing as a whole. This independent contextualization allows the budding developer to know when it is appropriate to use a new tool, whether or not a tool does its job (and how well), and what to watch out for in order to avoid bugs in dependent software.

Another crucial thing to being a good programmer is learning how to teach yourself. If a would-be developer goes out, gets a CS degree, and stops learning thereafter, they might as well be frozen in the year in which they graduated. New tools come out all the time, and it's important to know how to sift through documentation (rather than relying solely on tutorials), how to find good explanations of new concepts, and how to get questions answered without necessarily having a buddy or professor on hand who can definitively resolve the issue for you.

I think that your idea would be really helpful to organize the especially good resources on certain topics by increasing difficulty, but at the same time, I think that it could be a bit of a crutch. Learning how to self-teach sucks because it's hard, so a startup like you're describing could provide an easy way out that would catch up to users sooner or later when they need that skill.