This is interesting because, to me, programing is a deeply visual activity. It feels like wandering around in a world of forms until I find the structures I need and actually writing out the code is mostly a formality.
This is interesting because, to me, programing is a deeply visual activity. It feels like wandering around in a world of forms until I find the structures I need and actually writing out the code is mostly a formality.
I would describe my experience of it similarly, but wouldn't call it "visual thinking" in the sense meant in the article, where one uses actual imagery and visual-spatial reasoning. Indeed, I almost completely lack the ability to conjure mental imagery (aphantasia) and I've speculated it might be because a part of my visual cortex is given over to the pseudo-visual activity that seems to take place when I program.
I'm especially sure my sort of pseudo-visual thinking isn't what the article means by "visual thinking" because I also use it when working through "piles of abstract math", which I take to very kindly indeed.
Is your "wandering" of this sort of pseudo-visual nature, or do you see actual visual images that could be drawn? Very intriguing if the latter, and I'd be curious to know what they look like.
> Is your "wandering" of this sort of pseudo-visual nature, or do you see actual visual images that could be drawn?
They're like if the abstract machines you talk about in CS theory classes were physical objects.
For example, thinking about a data processing pipeline, I might see the different components performing transformations on messages flowing through it. I can focus on one component and think about how it take apart the message to extract the structure it's trying to manipulate, interacts with its local state, etc. If something is active and stateful it feels different than if it's just a plain piece of data. I run the machine through its motions to understand where the complexity is and where things could break, comparing different designs against each other.
I'm thinking about a data format, I think about the relationships between containers, headers, offsets between structures, etc, like pieces that I can move around to see how their relationships change and understand how it would work in practice.
It's more than an image that can be drawn because the pieces are in motion as they operate. It's the same kind of "material" that mathematical objects are made out of when I'm thinking about abstract math. It's immensely useful skill for doing my job, in designing systems.
I actually struggle a lot with translating the systems in my head into prose. To me, certain design decisions are completely obvious and wouldn't need to be stated, so when we all understand the product goals I often to neglect to explain why a certain thing works the way it does, because to me it's completely obvious how it's useful towards achieving the product goals. So that's something I have to actively put more effort into.
I also really struggled when I took a real linear algebra class, since it was taught in a very blackboardy "tabular" style which was harder for me to visualize. I was unfamiliar with it due to being used to thinking about matrices in the context of computer graphics and game engines.