[sbov]

I like both.

The problem I find with surface level knowledge is you make a lot of mistakes. When I have a surface level knowledge of a piece of tech and make a decision based upon that, then talk to someone with deep knowledge it's pretty common for them to say "why didn't you just do XYZ, it's easier/more robust".

Ultimately you can produce OK or even sometimes good solutions to problems but not great ones, unless you have direct access to people that know more than you.

This becomes even more pronounced when different pieces of tech can be combined with eachother.

[JimboOmega]

I simply do make a lot of mistakes in most domains of my life.

I'm ADHD (diagnosed, etc), and medication did help, but didn't solve it. For much of my life, I fought against this natural tendency. But for most part I essentially kept telling myself "pay more attention"! It didn't work.

Compensation is important. Detecting the mistakes I make, and avoiding situations where the cost of them is very high. One part of that is accepting things I can never do, like be an academic :).

In the engineering world there's lots of means of compensating. One you mentioned; knowing people who know more than you. Another one would be automated testing; at almost every workplace I've had, I have put a lot of effort into expanding and increasing it, and generally detecting errors wherever they might occur. I'm always asking "what if there's a typo somewhere, will we know, will that go to production?" Might I create a PR I think I've tested but I haven't? The odds I've done this are higher than my peers, so I need to protect myself from that.

If I'm at a workplace that values accuracy above else - that views being able to focus on the code and get the best (not "good enough") solution to every problem as the main trait that I'm evaluated on, I will not succeed. I'm never going to be the guru of anything. This kind of sucks, because a lot of places view being really good at those things as necessary before you can move to another area (IE, management) where you don't use those skills.

I'm a "good not great" or "move fast and break things" kind of person. My usual working style is to produce prototypes very fast; then whoever I'm making it for can see that I didn't quite read the design thoroughly and miss something before it's a huge disaster.

I need the feedback of fast release cycles. I can detect errors fast, take action fast, fix them fast. But "take your time and make sure it's perfect before you throw the switch"? That's just not my personality. I'm high volume (when not artificially constrained) but error prone.

I'm smart enough to know it, smart enough to insist that my code gets reviewed, smart enough to say we really have to test for these cases, things like that. But it's absolutely something I have to compensate a lot for.

[nkingsy]

I don’t have ADHD but I can relate to feeling that deep knowledge is inaccessible to me. My issue has always been around some kind of internal optimization function that is always asking “how will this help me”?

A solution that’s worked well for me is to only study deeply when I have a concrete need.

For example, when I felt my work suffering because I didn’t deeply understand z-index, it was easy to sell my brain on reading the complete z-index spec.

One of my hobbies is music, and I’m just starting to feel the limitations imposed by not understanding music theory deeply. That means it’s a good time for me to study music theory.

Some people feel compelled to learn things. I feel compelled to do things, and if I can’t do something as well as I want to for lack of knowledge, then I know I’m ready to learn things.

[JimboOmega]

Motivation is a super important factor for me. The gap between what I can accomplish when I care about somethin versus when I feel it is an obligation or requirement is huge. Thankfully the latter, combined with being smart enough to cover mistakes, has been good enough to keep me from failing things or getting fired, but it's a definite challenge.

Confuses the hell out of employers, too. "You did a great job on X, you really improved things, but when we put you on Y, you totally halfassed it" - because X was my idea, and Y was forced on me. I respect the (social) contract that you are my employer and I should do what I'm told; I know at least mentally I should try to excel on any task given me, but it's just so hard to invest in those kind of things.

Anyone can get demotivated; it's just that the gap for me is bigger than people are used to. I'm not blaming my employers here, to be clear - thinking that way does me no good, anyway(+). While the vast majority do a terrible job of recognizing the motivations of employees - and anyone can get demotivated - I'm going to be way worse at this than the average employee and it's up to me to compensate.

Another strategy: I learned early on ANY social context is motivating for me. So I try to schedule regular checkins, even if the org doesn't mandate them. Even a few minutes of a manager explaining the next project will yield much better work than getting a ticket thrown over the wall. Constantly engaging with others and feeling like they are personally depending on me is one of the most motivating things out there - so I have to keep reminding my bosses that YES, I do want to be in all those meetings.

I will say on the subject of learning - I mostly agree with you. I learn all the time in a huge variety of fields, and when I need to learn new tech to accomplish a task I relish it. But if I try to make myself learn something because it would push my career ahead? Force myself to devote some number of hours to a project in a new language, something like that? It's like pulling teeth. I can't get into it.

(+: In general, if the world discriminates against you, is difficult for you, etc - whether or not it's truly worse for you or a perception you have, it doesn't matter to you - what matters is how you react to it. Sitting at home crying how about how it's not fair will never benefit you.)

[spiralx]

Thanks, as someone else with ADHD this is almost entirely spot-on with regards to myself, although I'm perhaps overly anal about the code I work on as a defence mechanism - which sometimes leads to focusing on the minutiae and missing other issues :)

> I need the feedback of fast release cycles. I can detect errors fast, take action fast, fix them fast. But "take your time and make sure it's perfect before you throw the switch"? That's just not my personality. I'm high volume (when not artificially constrained) but error prone.

So true, and it's made past jobs very hard work, most especially before I was diagnosed at age 31 and had the benefit of medication and being able to formulate better coping strategies :)

[zhoujianfu]

Me too.. in fact I just started imposing on myself last week a MWF: on top of things, TTh(plus weekend if available): on bottom of things. Hence here I am on HN on Monday!