When I was a little kid, I had a very strong (and I now realize, unusual) affinity for the truth. I would feel bad if I told someone a fact which I later learned to be false. I felt that if I said something that was false, I was contaminating people's understanding of the world, and this was a wrong thing to do. I didn't frame it this way, in those words, of course, but I remember that this was my explicit thought -- that spreading inaccurate representations of the world was a bad thing to do.

I wonder if many hardcore technical people (I am getting tired of the word "geek") perhaps share this early attitude of mine.

To indulge in some evolutionary psychology, I wonder if this aversion to spreading misinformation was an adaptive trait back when proto-humans lived in small troops -- spread misinformation, and people on whom you relied would be lead astray, to your own detriment.

"I want to say true things, but apparently other people want to hear lies," I thought to myself. But I now know that it's more nuanced than that. Typical people ("non-geeks") do care about truth when it directly relates to what they're interested in: usually social status and mating, and if they're unfortunate, survival. Generally, they're not interested in more abstract subjects.

I interpreted people way too literally. If they talked or asked about X, I assumed they were curious about X. I didn't grasp that small talk is about entertainment and insecurity: the speakers reassure each other of their companionship; gauge each other's abilities, motivations, and desires; signal tribal affiliation; share gossip; and determine their places in the pecking order. It has nothing to do with X. X is just a pretence.

Even as I became conscious of this, I often acted the same way out of principle. Alas.

I think that it is even more nuanced than you propose. Often non-geeks are interested in the subject, but may not have sufficient interest or mental energy required for significant, structured thought into the subject. That is the kind of thought that geeks are accustomed to, but to non-geeks, that's what they do at work, not for pleasure (exceptions abound, though. For example, "non-geeks" often put careful thought into sports. Maybe they should be called "sports-geeks").

I regularly am asked questions about technical matters outside of my expertise. I try to draw some logic into the matter, break it down, and come up with a simple, reasonable, and probably accurate answer. If required, the asker is often intelligent enough to come up with a similar answer - or at least an equally plausible one. But they don't want to, so they ask me. And that's okay. If I don't want to think about, I'll tell them I don't know.

And it's even more nuanced than that. Someone might be very interested, on an absolute scale, in some subject. But as long as he's even more interested in social reassurance, gossip, networking, status, and so on, he will prefer conversations that revolve around those instead. It's a matter of priority. Yeah, there's a latent physics geek in his brain, but when he asks you about next week's weather, his goal is imprecise small talk, not a discussion of turbulence models.

At first I felt like I was being wishy-washy (changing my tune) but I later understood it as just an attempt to be right and not give out wrong information. I felt giving out wrong information and not correcting it when I found out it was wrong was breaking the "trust" relationship I had with the people involved.

Even though I am still in that type of a position and still delivering those types of answers, I now attempt to couch said answers in language that shows "this is what I currently believe, and this may change if I get contradictory information."

Also, I notice I am much pickier about correctness if I have recently been programming. My wife likes to point this out: "Oh, you're in THAT frame of mind."

I guess no one got the obscure Feynman reference.

For example, A geek will know various properties of PI out to some number of decimal places etc. If they have to calculate with PI they'll try and use as much precision as possible in order for the results of the calculations to be as correct as possible at the end. A geek knows that if they just say "PI=3 is close enough". Then calculations using PI end up with all kinds of weird results...and cars will fail to roll and houses will fall down, and ballistic trajectories will miss their mark etc. So a pedantic fetishizing over the digits of PI becomes a necessity. It's hard to turn on and off that level of mental discipline, so by force of necessity most geeks simply set the switch to "on".

So a geek necessarily internalizes the practice of being pedantic and sweating the details.

This happens with anybody who sweats the details in a narrow field because they understand that you can reach the wrong conclusion at the macro level if you don't. Even in seemingly traditionally un-geeky fields like fashion or cooking, you find similar types of behavior patterns.

What this essay is identifying is either the lack of the ability in technical geeks to turn off the pedantic-ism or to not forcibly apply it to all aspects of life.

(To me, this is the question that needs answering.)

A few quick illustrations off the top of my head:

* I speak Spanish fluently and was traveling to Honduras with a group of twenty-some other Americans, only one of which had similar competence with the language. They asked me to give them a quick "Intro to Spanish" course. I started with a simplied academic view, "I'll start with some simple vocabulary, some verbs, some conjugations of 1st/2nd/3rd-singular/plural forms to give them an idea of how to talk." I quickly abandoned that, stuck to the "where is the bathroom?" kind of sentences, names of common foods, other words, and a little human-interest history/geography. That approach worked, they could ask for bananas even if their word ordering and conjugation wasn't perfect. They would have been bored to tears after two minutes of a formal language presentation.

* I was working at a startup and a cute phone-sales rep wanted to know more about our product so she could sell it better. She enjoyed biking too. Great, we go on some great bike rides, I try to educate her about our product but went into way too much depth. Her eyes kind of glaze over, I'm really disappointed about her lack of interest in the product. "You really need to know this," I tried to explain. She probably didn't. She just needed the talking bullet points. She wasn't interested the engineering world view of things and I couldn't deal with that. I saw some disturbing rigid tendencies in myself too. The friendly relationship kind of stopped at that point.

* The best support person I've ever worked with was a master at glossing over the truth, in a good way. He'd talk to us engineers, we'd tell him exactly what was going on and how we needed to change a configuration or fix a bug or apply a patch. What he told the customer often had nothing to do with what we'd tell him. And we and the customers really were better for it. Either the explanation was too detailed for the customer to follow it, or would alarm the customer and bring up a whole bunch of questions they really didn't need to be asking. That's wisdom, of a sort, at least.