When starting a new project, how can you know what tools to use, how long it will take, what roadblocks you will run into, or if your method is even close to the most effective way of solving the problem? I’ve been running into this between our research project, and also my deeper dive project recently.

I think this is probably why having some sort of mentorship is so important. Finding a mentor who shares similar interests as you and understands where you are at in your knowledge is so helpful because they can tailor help specifically to you.

I think it’s really weird how I have realizations like this, even though they seem so obvious in retrospect. But I guess it’s just really hard to listen to advice from others if you’ve never had to deal with the issues yourself. (this seems like some fundamental limitation of humans, which is really interesting)

I have kind of been thinking about this in the realm of programming lately. Sometimes to appreciate why one complex method or tool is better, you have to do something the simple way first, and figure out for yourself why it sucks. If you had listened to someone’s advice, and gone straight to the complex solution, I feel like you wouldn’t have understood so many parts of what makes it useful. (e.g. I didn’t understand why Anaconda was better in some scenarios compared to venv until I used venv for a while and ran into a lot of the issues that Anaconda solves)

Don’t read the rest of this if you’re bored, it’s more of an unstructured rant

but I feel like what I’ve written can’t be a hard and fast rule, because there are so many modern tools that 10 years ago would have seemed like the very complex soultion, but now they are the first thing that anyone learns. It’s such a weird feeling to go into learning something like programming where, when you are first learning it, there are SO MANY unexplained things. And yet you are expected to still use the tools to create things, knowing that you don’t understand why certain behaviors exist. I feel like this makes creating real world tools outside of a classroom very difficult. Would it be better to teach from the ground up? Idk probably not, but I feel like the current method is werid.

Maybe the reason I feel this way is because I didn’t have a mentor for a lot of programming experiences? Which actually validates my rule?