How I'm teaching myself math slowly
Growing up attending public school, math was without a doubt my worst subject. And that's always bothered me.
Especially as a programmer. I can build incredibly complex systems (that do math themselves), but on paper, I struggle with basic concepts. This has limited—to some degree—what type of programming tasks I can actually accomplish, and so, I've decided to shit or get off the pot.
Instead of doing the old-fashioned approach of solving problems and reading books about math, I've decided to leverage LLMs.
There's a book on my shelf A Programmer's Introduction to Mathematics that's gripped my attention in the past because it helped to reframe math from the perspective of someone writing code. There's a lot of conceptual overlap between math and programming, so I figured that's the easiest way to go about this.
To start, I decided to ask ChatGPT to put together a "course" for me, framed to make sense to a programmer:
I want you to teach me math. I've never been great at it due to how public schools teach/taught math. The only time I got an A in math was in college algebra I. I just sat down in the cafeteria at college and taught myself from the book. We want to frame this from the perspective of a programmer. Let's start with just general "important to understand this thing, what it is, why it exists, and how to apply it (if you even can)."
Conveniently, the response I got here was pretty helpful right off the bat. This explanation in particular really helped to unstick my brain and reduce fear around re-learning math:
As far back as my memory can take me, I've never heard anyone explain it in this way (that's not a claim that no one has—of course they have). It makes so much more sense to think of an equation as just "two sides of the same scale being balanced."
This got me thinking. I replied to someone on Reddit the other day about learning C++ and using an LLM to reframe teaching it to them by reframing it in a language they already understand. This is how I've dabbled with stuff like Go and Rust in the past, so I knew this approach (generally speaking) would work.
I share this not to brag about learning math (I can't think of much that's less cool than that, so...) but to suggest that learning doesn't have to fit into the models we're conditioned to believe it must. It makes a ton of sense to have a more personalized, contextualized version of learning. It makes you less avoidant and makes the information easier to both digest and memorize (because it's framed from an existing world view you're presumably comfortable with).
I've tried to relearn math several times now, so I can't really say if this gets me where I want to go. But I do like the sort of self-paced, personalized way of learning this stuff. I'm not big on cutting out other humans, but when it comes to learning, this is one of those places where doing so can help more than hurt (up to a point).
Having taught others in the past, I would have preferred if I could have mimicked an LLM. Instead of trying to figure out one-size-fits-all explanations, on-the-fly tailor whatever is being taught to the individual. Not in the "fewer students per classroom" political sense but more so in the private tutor sense.
There's something here when it comes to learning, but I'm resisting my typical urge to codify it into a product/solution.
Either way, I'm relearning math. I have a strange feeling it's going to come in handy in the not-too-distant future.