Thinking, Coding, and Writing: An accidental post

Rob K. Henderson is a really smart guy. If you’re on twitter go follow him: Either way go to his website and sign up for his newsletter and read his stuff:

He posted a poll on twitter ( asking “Close your eyes and imagine an apple. What do you see?” with the following results:

A vivid image of an apple 45.5%
A fuzzy image of an apple 21.9%
Image resembling an apple 14.2%
Nothing/blank 18.3%

I’ve seen descriptions of this experiment before and the results are always fascinating. The comments underneath are also pretty damned enlightening. They of course run the gamut between “Wait what? There are people who actually…like…SEE images in their heads?” and “What do you mean you don’t? How could you not?”

And it really does make perfect sense. But we assume we all think the same. At the most casual level we assume we have the same underlying principles and values. Sometimes people are shocked to find out that we just don’t. Still stranger is the awareness that the toolboxes of our brains are wired differently at a deep biological level.

I wonder a lot about that experiment and what it does and doesn’t mean. Where the lines REALLY are and what they mean. Because I suspect that, while there really IS a pretty concrete difference in the deep underlying mechanics of how we think, maybe a few different dimensions that we all fall somewhere along, like…oh I don’t know, maybe how abstract or concrete our mental “images” are, how we process mimetic information when it comes in through any of several sensory channels. (For instance, I learn best by hearing words rather than reading them.) And maybe there are a bunch. I’m no evolutionary psychologist or whatever the hell that would be.

But there’s an equally large variance in how we conceptualize things and therefore, how we actually describe them. Do the people who say “image? what image?” assume that there’s a literal hallucination involved? Stated another way: To what degree is it a difference in internal representation and to what degree is it a function of available language and taxonomy? To what degree is it just a different “way of talking” either through education and articulative acumen? I suppose some Smart Guy could come up with a battery of texts to figure that all out.

I don’t know. I can’t quite see it. cough

For my part I don’t get an “image” of an apple. I get what I can only describe as a conceptual construct. I can rotate it in space, almost like it’s a wireframe of an apple. I have a perceptual sense of structure that’s not “vision” in any real way but is definitely “spacial.” It’s actually what I imagine goes on in a blind person’s head when they imagine an apple.

One of things I find really interesting about all this is that it’s how I conceptualize and design software.

When I’m working on a piece of software in my head I absolutely represent it as something close to “objects without form” in conceptual space. They only have “shape” insofar as they have to interact with each other. But they have size, distance, and relative position. I think about things and how they fit together by creating a mental model (one of the most overused digrams in the English language in the 21st century) of the components or functional units and see how they line up.

But again, there’s no “image” I could create from it that would be an accurate translation of what I ACTUALLY imagine.

Now what I get out of doing that is the ability to create hierarchies of models. This thing represents a log file or a network connection or a record in a contact management database. Then I figure out how a few little subsystems relate to each other and create a higher level ‘mental object’ out of that. NOW I have an object that represents a higher level of abstraction. So I can easily bolt THOSE together without having to worry about all the damned underlying details every time.

Models built of models built of models. Do I have all the details in my head at all times? No, of course not. That would take a truly incredible amount of mental processing power. BUT I can trust that if I have a conceptual model of a thing, that I can drill down in to it at will, should I need to, to see how it actually works.

On one hand I’m rederiving the guts of it from scratch. But I know, because I don’t build a higher level model but that I do so from lower level models, that whatever is under the hood, so to speak, is something that I can get back to with relative ease.

At that point actually going from model to code is just a matter of typing, for the most part. The stronger and more thorough the model the less work is involved in getting it in to code.

Now this works equally well from the top down. “I need a thing that does a thing. Now…what are the moving parts of that thing? What HAS to happen in order for that design goal to be fulfilled?” Then I break that down into smaller and smaller objects until I get to either components I’ve built before or things I have to research, those strange little clouds of “I’m not sure how to deal with this but I can just go look that up someplace.”

One of the most useful aspects of this process is that if there’s something I don’t understand, a technique or tool I can’t see clearly yet, then I can say “well, there’s something that sorta looks like this from the outside that goes here and I’ll figure out the details later.” So I don’t REALLY need to understand the whole thing before moving forward.

I can draw images approximating what I think in something like UML (UML is essentially a diagram style for representing software components and their relationships) even though it requires a significant amount of translation. Because it exists in “form without form” it doesn’t get all wrapped up in the “object” metaphor, but I’m still able to manipulate it all as if it were.

It’s a (well…THE) technique I use to deal with just about everything I think about. A buddy of mine wanted me to look at his 3d printer and troubleshoot it. I…don’t know why he thought I had any particular insight into 3d printers. They’re just glue guns on a gantry. But I took it home and sure enough, it was behaving funny. So I took a look at it and spent some time thinking about what the parts are, what they’re doing and what MUST be going on and I started bifurcating the problem. I’d figured it was probably something simple because the whole thing is just a composite of simple machines. No, I haven’t solved the thing yet. But it can’t be TOO hard.

Now it’s not all hookers and blow.

One of my biggest frustrations with writing (well, almost certainly my biggest) is that I haven’t figured out how to model words that way. I can’t seem to take a piece and break it apart into conceptual abstractions of individual units in order to structure it at a higher level. As a result everything I write comes out as a kind of stream of consciousness with whatever structure I can apply as I’m going forward.

It shouldn’t be this way, and maybe it’s a function of personal training. I get the impression that I should be able to do this. Break a piece down from top to bottom, see how the components interrelate with callbacks, arc of progression and such and then drill down from there to put it on paper.

But that’s almost the exact opposite of how it works (or, well…how I work.) I just start typing and let the structure emerge out of the nominal chaos that comes out of my fingers. Then once I’ve gotten to about 600 or so words, I start thinking about wrapping it in a beginning and an ending (assuming I didn’t start with either, which is what I actually do most of the time)

I suppose what I really need to do is start forcing myself to conceptualize of writings from the top down and really organize them in the same way. I don’t see any reason why it wouldn’t be able to work. But I just don’t have the tools in my head, the templates and models of what these things should really be in order to build more cohesive pieces that actually are able to approach the definition of “essay” if nothing else.

And, as an object lesson in the phenomenon I’m describing…I have no idea how to end this, where to take it from here.

Okay here. I’ll do the cop out ending.

What about you? How do you answer the apple question? How does that come in to play with your thinking? Does what I’m saying here seem self-evident or completely alien?

I know it’s a cop out ending, but I actually AM curious.