Read on Twitter
A random thought: mentees and mentors in the sciences and computational fields need to understand the X-Y Problem. I help a lot of people with computational stuff and early on they'll often hear me ask "wait, what is it you're trying to do... is this an x-y problem?".
An x-y problem is when someone comes to you with problem (usually computational) asking you how to get y working. A clever helper will immediately ask "what is it you need y for exactly?" and find out the user is actually trying to do x, and thinks y is the solution.
Far more often than not, the right way to do x is *not* y — e.g. there's some function they weren't aware of, or some API, that would make solving x far easier than their solution y. I would say 80% of the computational problems I help with are x-y problems.
