This is a far belated reply to this thread, but I would introduce this thought. Stepped output of any sort is really a philosophic question about SAGEs direction. While yes, SAGE is a wonderful hodge podge of CAS and other elements, it's not really in my mind a 'tool for checking homework'.

If one wishes to implement Stepped output for various calculations, the reason and roadmap would need to be clear, as this would be really time consuming (I've found that even wolfram alpha doesn't clearly demonstrate the order of operations in evaluating variously written expressions).

If a paid application, where resources are directly allocated to its development, isn't getting this entirely correct, then there are likely some hidden gotchas in displaying symbolic output. If you've ever written anything in LaTeX, you'll probably get what I mean by this.

Still, as a general idea of SAGEs direction, would SAGEs developers at UW wish to consider branching SAGEs platform into one which has features for educating in maths, etc? This would probably be an enormous project, but given Python is the primary underlying primitive and given dev time associated with projects (vs C based) , I wonder if development in this direction can be proposed, planned, and 'crowdsourced'.

It's kinda daring. But, SAGE is extraordinarily robust, and performs nicely for scientific computation (in fact, I think it would be worthy of a performance test in the Julia circles, given what very large components underlie SAGE). I think the 'extension' to various educational circles is worthy of consideration. I'm certainly willing to put my time into this. Thoughts?

The way you solve a problem by computer and by hand is very different. Do you want something like Mathematica's `Trace` which only shows the (non-kernel) rules that Mma applies to an expression or do you want something like the "Show steps" in W|A that is an extra layer and a completely separate process to the way the underlying program works?