Update packages used

[KronosTheLate]

I can see that in https://docs.juliahub.com/CalculusWithJulia/AZHbv/0.0.13/precalc/julia_overview.html you use

Pkg.add("SymPy")  # for symbolic math
Pkg.add("Roots")  # for solving  `f(x)=0`
Pkg.add("QuadGk") # for integration
Pkg.add("HQuadrature") # for higher-dimensional integration

Please consider the alternatives detailed in https://discourse.julialang.org/t/julia-for-education-package-selection/82852/2 that are part of the SciML ecosystem, which act mainly as uniform frontends to any number of under-the-hood backends.

Specifically, Symbolics instead of SymPy (unless you need symbolic equation solving and symbolic integration - those are still missing), NonlinearSolve instead of Roots (I actually also really like Roots, but perhaps mention NonlinearSolve), and Integrals instead of QuadGk and HQuadrature.

[jverzani]

Thanks for the input. I have plans to write up notes on these alternatives this summer, as I have looked at that thread too. I am reluctant to shift to Symbolics, as the calculus specifics (limits and integrals) aren't as competitive, as well the non-linear solving. But it will improve and at some point that shift will happen. Integrals is really new, but that could probably be incorporated now. As for Roots, I'm kinda partial. The alternative is faster (but not so much so) and has advantages, I believe, when there are multiple starting points, but I'm not sure the extra complexity is necessary. But I'll have another look over the summer.

[KronosTheLate]

Excellent. I mention this because I accidentally came across your result when googling, and thought that the quality of the packages suggests therefore is important as it could guide new users. I agree with all your considerations, so I have confidence that you will strike a good balance after the summer ^_^

I will leave this issue open for now, and you close it whenever you feel that the currently suggested packages are a good pedagogical selection from the ecosystem.