#Brown Book forum feedback

11 messages · Page 1 of 1 (latest)

junior onyx
#

Hi folks, I'm the main author of Brown University's augmented version of the Rust Book. I saw that some folks here had thoughts / concerns / feedback about the book. Happy to answer any questions about the design decisions behind the book. Also happy to discuss any changes that y'all think would be valuable.

cc @dusty dome @woven sage @flat rock

flat rock
# junior onyx Hi folks, I'm the main author of Brown University's augmented version of the Rus...

Thank you for the openness and transparency! I appreciate that!

I haven't read the book personally, so I apologize if there's not a lot I can say about it (certainly not as clearly as others who've actually read it).

I can answer a few concerns I've seen, but I don't have much on the feedback front.

(Note, if some of what I say is inaccurate or even incorrect, please look past it in light of the fact I haven't really read it; also, the last time I read the rust book was 6 years ago)

For concerns, the main sentiment I've seen is:

  • I think I saw some discussions before about some of the quizzes may be either incorrect or partially incorrect on some fronts (I apologize I can't say which, but others may be able to chime in here; I think it was about a UB quiz; though on the other hand, you know how UB is and how internal details can change)
  • the "big" one is often that people get real confused about vs the main book is the RWO system of explanations. When they read the official book, it generally gives explanations which seem easy enough to understand and wrap the head around (aliasing XOR mutability). * But when I read these, it left me a bit confused (a lot of cognitive overhead)
  • From what I understand, there wasn't really much other than these 2 points I guess. Though others may have more to chime in on regarding this as I've, again, not really read the book

Some things I loved compared to the official book:

  • aquascope. Seriously, this tool is pretty cool, and I personally feel the diagrams can be real helpful. While the main book has some of these, it is a bit light in this regard.
  • I absolutely do think the quizzes are a good feature to test people's understanding / help them think things through.

* note, I can't speak for everyone, so it is possible it is helpful to some or many people; this might need some data to verify. I'm only going off of what I personally felt / have seen

#

Feel free to ask me if you have any questions about the above; I'm happy to answer to the best of my ability

junior onyx
#

This is a helpful list, thanks. I appreciate the positive thoughts. Re: concerns, I am planning to revamp the quizzes that talk about UB to try and quiz the same concepts while making the wording less controversial. To me the more concerning thing is whether people find the permissions / RWO explanations accessible or confusing. The goal is to give people a mental model for how the borrow checker "sees" your code at a level of detail greater than aliasing XOR mutability. In particular, we want to help readers distinguish between Rust code rejected for good reasons (e.g., dangling pointers) and bad reasons (e.g., disjoint mutable indexing), even if the compiler errors look similar.

#

Do you have a sense of which elements of the permissions system that Rust learners find confusing? Or is it more a general sense of some people try reading it, and just don't "get it" broadly?

flat rock
#

If I do end up rereading this section so I can give some feedback, I'll keep what you've just said in mind and read it with that intent ferrisCat

junior onyx
flat rock
dire nest
#

👋 you have already seen my take at length, but I am also available here if desired

bronze goblet
#

The diagrams in your book add alot of value for me, and when I recommended people start learning rust I generally send them to your version first and the official version second