I do not like to use homebrew for python, but I install it for Homebrew dependencies.<p>I use the native installer available on python.org. It works perfectly on my mac, and gives me no issues. I create local virtualenvs for each project, from the base distribution of choice.<p>I have found zero reasons NOT to use the standard distribution, and it allows me to install any version of sub-version I desire.
I can't recommend using Anaconda/Conda it's so bloated.<p>A poetry/.toml/requirements.txt file and venv is enough to manage without having 100's of data science libs installed.<p>> Feel free to use conda but keep it scoped to individual projects rather than as a system-wide configuration.<p>Good advice!