[trevor-e]

> And I love that the problems remain difficult and meaningful

And there it is, your argument comes down to gatekeeping.

If all the problems have been publicly solved, as you said, then what's the harm of having a solutions guide? The cheaters can already find what they're looking for and it helps the self-learners who are actually trying to learn.

[jefferickson]

> And there it is, your argument comes down to gatekeeping

Bullshit. If you want to play basketball in the NBA, you have to practice your ass off to develop the necessary skills. If you want to be a successful car mechanic, you have to practice your ass off to develop the necessary skills. If you want to be a successful cook, you have to practice your ass off to develop the necessary skills.

The homework is a vehicle for practice. Effective practice is real work. Real work is hard. Therefore, the homework must be hard.

[aspaceman]

They haven't. A lot of the exercises are unsolved research problems.

You're missing the point entirely. You're just banging your head against "Different people learn in different ways". I am saying, that the thing this course, (that I took by the way), is trying to teach students is the skill of writing proofs effectively.

A solutions guide, as you imagine it, would not teach students to write proofs effectively. Since many of the problems in this book are of a research level, many are highly advanced. Many require many multiple pages of exposition to describe the solution effectively.

You would thus have to make a tradeoff. Remove the harder questions and write simpler explanations, so you can write a cost-effective and printable solutions guide. But then the "solutions" aren't actually solutions at all, but vague allusions towards the correct process. The student assumes that the "answer" are these vague allusions, "since that's what the solutions guide says" and the entire point of the class is bypassed in favor of passing more students.

It has absolutely nothing to do with gatekeeping, or cheaters. And quite frankly, I don't believe that advanced mathematics will ever able to be the most accessible thing in the world. And that's just its nature.

EDIT: Seriously if this was Calc, I'd totally agree with you 110%. A lot of people can only learn that stuff by staring at the answer being worked out. But that's not math man, and it's not what we want to teach people in this course.

[trevor-e]

Thanks for the more reasonable take, though I still have to disagree.

What you're saying directly contradicts the professor's own statement:

> Please do not ask me for solutions. With very rare exceptions, I will say no, even if you are an instructor. I recognize that my stance limits the utility of these materials, especially for self-learners, but I'm trying to optimize the learning experience of my own students at Illinois. The point of homework is not to solve that particular homework problem, but to practice solving a type of problem and get honest feedback on your progress. I've found that when solutions are available, my own students are much more likely to rely on them, rather than trying to figure out the problems themselves, which means they get both less practice and less honest feedback, which means they do worse on exams and in the course overall.

> And while I firmly believe that each student is ultimately responsible for their own learning, I also believe that it's my responsibility as an instructor to help them. Putting dessert on the table does not help anyone eat their vegetables.

tl;dr: students tend to rely on solutions when they are available, they don't learn as much and get worse grades which reflects badly on him.

It has _nothing_ to do with a tradeoff of writing simpler explanations for a more printable guide. He even acknowledges that it limits the utility for self-learners, which you disagree. He's a professor with obligations to his students so I understand his decision, just sucks that an amazing book won't be more accessible to people.

[codeisawesome]

Thank you for engaging the other commenter and articulating my stance better than I could have. Also - while I have deep respect (I truthfully do) for the expertise of folks who teach this material - I have lost patience for their patronising tones and unhelpful assumptions about the folks that might read this book and books like it.

The world needs an algorithms book that shows a modicum of humility to its readers. A book that assumes that more than simple children are reading it (incidentally, and in a puzzling contradiction - also using needlessly convoluted English to express ideas). Hopefully I can do something about that ‘one day’.

[trevor-e]

Totally agreed, the academic world has always operated and depended on a walled garden approach and making things needlessly complicated. Thankfully the cracks are forming, the pandemic probably sped things up quite a bit with universities trying to charge students to essentially watch Youtube videos.

What's funny to me is that UIUC offers an online CS masters program now. Not having accessible solution guides seems like a disadvantage to these students.

[aspaceman]

Let me write these solutions to these research level problems....Just give me a couple years and some funding please. They don’t have solutions yet. Hello? You have to figure them out because the professor doesn’t know.

Are you even reading what I’m saying? How do they write solutions? I feel like a crazy person as everyone is talking about a walled garden. Like hello? Do you understand what, “there is no solution until you find it” means? Do you understand what, “there are many worked solutions in the text” means?