But at the same time I acknowledge there have been strands of progress, and for me personally (even if this is not the vision of the project) it has pedagogical value to understand and convey to what extent a new contributor can use "the usual tools" to develop SciPy, avoid deprecated workflows whenever possible, and in sum set boundaries to "SciPy exceptionalism" with proper explanations. Then of course most people will not want their head to explode and say "give me a workflow that works", but as I said, I'm trying to go a bit beyond that, and this thread has been very illuminating in that respect.