If you want documentation and comments, to start with you should stop calling them programmers.

If you want documentation and comments, you call them software engineers and you allocate time for documentation, commenting, optimization, debugging, and other critical non-functionality tasks. Managers typically only allocate time to develop something that sort of works, not something that's polished, maintainable, debuggable, extensible.

The gripe is that some students are less prepared than others, making it difficult to design courses? You're always going to have that problem. In the third-semester CS course I just finished teaching, there were 25% CS majors that took the prereq in Java, 60% or so CPEG or EE that took it in C++ two semesters ago, and the rest took prereqs at community colleges or abroad. Depending on what they took and their background, they were unprepared or overprepared. You can't teach to the least common denominator; you end up balancing the curriculum so that it's sub-optimal for almost everyone, but equally sub-optimal for many students.

The music analogy is awful - it's reasonable to expect prior music experience for a music major and unreasonable to expect them to place well in the orchestra right away. The "CS is not programming" comments aside, maybe they don't have any background due to their highschool (although that shouldn't stop a hobbyist). So you solve the problem by having an "intro to programming" first semester and placing many students out of this course (that's what happened to me in undergrad and it was fine). Or you have them take remedial courses first because they're behind.

Getting back to CS1 - passing the course doesn't mean you can program anything. Depending on the school, it means you can write toy programs, typically with awful readability/performance and typically it doesn't indicate that you understand recursion, pointers, complex data structures, etc.

The most important thing is that every school is different and struggles with different problems. Some schools prepare you for a PhD in CS, some schools prepare you for the workforce, some schools prepare you for nothing at all, some prepare you just to learn and deal with strict requirements, and some mix a little of it all. It's absurd to try and make a general claim about computer science in general.

It costs so little and can really boost productivity for development. Then again it depends on the person. Once I went to dual screens my productivity increased and I can't really deal with single-screen for most tasks anymore. The bigger question to my mind - if your company doesn't value a simple $100-200 investment in developer satisfaction and productivity, how do you expect to retain your developers or maintain a productive team?

The title should probably make the distinction between masters and PhD students. Either way, I can't help but feel like the question is better left unasked. The vast majority of the PhD students I've worked with (myself included) have had a lot of hard times in their program. And there's a vocal subset of those that seem to almost compete in terms of who feels the most victimized. That's not to say that the system is fine; there are significant issues in the US PhD system. But if you posed this question to PhD students over drinks, you probably wouldn't get a meaningful answer. There isn't really an easy way to compare the issues in biology to computational chemistry to theoretical computer science to applied computer science.