Ask HN: Why are expert practitioners often so bad at teaching beginners? | Hacker News Reader

Ask HN: Why are expert practitioners often so bad at teaching beginners?

30 pointsnewsoul36 comments

This may be only my observation, but, I have observed this many times. A truly exceptionally expert practitioner doesn't always know how to teach a beginner the workings of his field. He knows his field inside out and are almost always self-taught in achieving their expertise. Still when someone asks how do they self learn (atleast start) they are clueless or point to obscure or very high-difficulty resources.

Is it because the area they practise in now, is at a much higher level of abstraction than what a beginner should do and they have lost touch? Or is there some other reason?

E.g: You ask an expert programmer for advice on how to learn to become a good programmer, they will almost always suggest mid tier resources.

View on HN

I think teaching is a skill by itself. Effective teachers usually grasp the theory behind the learning process, while this is not necessarily true for expert practitioners. There is much more to this: knowing how to communicate effectively, how to keep learners engaged, adapting to your students etc. This means a person can effectively teach the basics without having deep technical expertise, it's simply not the most important skill in teaching beginners imo.

So assuming those are somewhat independent skills, you can find skilled practitioners and skilled teachers but in my experience it's hard to find someone with that skills intersection. I consider this kind of people quite exceptional.

I agree. I have gone from working in tech to tech teacher.

One of the hurdles in teaching, over being the office tech wonder, is when the penny drops it's not all about you.

Because complicated things become so obvious and natural that it's hard to even remember that some things are not inherently known by any of your peers.

It can be hard to accurately gauge your interlocutor previous knowledge, so you don't make them feel dumb by explaining too simple things, or by starting with too complex stuff.

>or point to obscure or very high-difficulty resources.

Although here I think that's on the learner to make an effort. If you want to learn X but you need Y to understand it, your next question should be how can I learn Y, and if you need Z to learn it, you should ask how to learn Z, and most of these questions can be answered by search engines, you don't usually need an expert for that.

"E.g: You ask an expert programmer for advice on how to learn to become a good programmer, they will almost always suggest mid tier resources."

Well once you went through something like CS50, you should be able to grasp mid tier resources, or at least be able to find out what you need to learn to go through them.

Context is often missing. Or basically, the four quadrant. People often talk about the known knowns, known unknowns and unknown unknowns, but most people forget the ( possibly most important one ), unknown knowns. And because the expert practitioner didn't even known they know ( something that they think is obvious but didn't know others have no idea ), it is hard for them to teach.

Unlike in Universities, where each student are given the same resources for them to read through and progress, most of the teaching in real world ( non-academic settings ) does not ofter the same path.

And without those foundational knowledge, a lot of discussions would be broken because the assumption of you knowing is untrue. For example, half of the software developers have absolutely zero idea how computer works. But they somehow think they do, as witness in the recent HN discussions. This makes teaching and discussions on a subject enormously difficult.

Edit: OH this is great. I just discovered there is a term for it in the comment below. The curse of knowledge. https://en.m.wikipedia.org/wiki/Curse_of_knowledge

I'm a strong believer in experts tend to forget what's it's like to be a beginner.

I was sitting down to join a voice call with a friend with the intention of introducing him to programming basics. I explained that we were going to "simply go over printing" and he informed he he wouldn't be able to work on this today, as his printer was in his basement and not accessible on such short notice.

Because to be expert the axiomatic building blocks must become instinctive, so you can use your limited working memory on higher order problems.

It's hard to explain instinctive things, or, for some, to understand why they are not automatically instinctive to others.

Sports is a good place to look for models since the hiring and firing of coaching staff is reported right alongside the other narratives during a season.

It's rare for the best coaches to have been the elite practitioners of a sport. What's very common is for someone who struggled to even break into the elite league or had a very short career there goes on to work through the coaching ranks and become an elite coach.

I have always thought it's due to mid-tier players needing to really focus on every lesson they get, studying every drill, obsessing over every piece of advice they get. Elite players just absorb the drills and lessons into who they are and forget it was ever taught to them in the first place. When the learning comes hard the player will understand the complete process to come to understanding. They will also see the elite players around them apply the lesson and improve, so they know the lesson works. Being worse at doing the skill can ultimately make you better at teaching it because you are forced to understand the skill more deeply.

Being a replacement-level player doesn't automatically make you great coaching material though, that's where the filtering starts to find great teachers and communicators. As many others have commented, teaching is a skill unto itself, one not everyone has.

This is a really key point. While teaching is certainly a separate skill from doing, there is also the fact that the top performers in any given field may have never truly been novices in the way most beginners are. Think of Andy Bolton deadlifting 500 pounds on his first ever attempt. What is he supposed to tell someone about how to do it? He never struggled, never had anything to learn. It's why Michael Jordan and Kobe Bryant were such terrible teammates. They just got frustrated that the game didn't come as easy to other players as it did to them and yelled at them for being shitty.

It's pure statistics: a really good teacher is rare in its own right, a truly great expert is similarly rare, and now you are looking for an intersection between the two.

Add to that other circumstances like no teaching education for most experts in other fields, and there's your answer.

They are cursed: https://en.m.wikipedia.org/wiki/Curse_of_knowledge

Well like you said, expertise is generally self taught so they don't really understand the model of someone asking how to learn. They never asked someone how to learn. They pick a goal/problem to solve (or have one thrust upon them) and churn through whatever books or googling it takes to learn enough to do it.

idk i think the same pattern applies in many fields. as a beginner you worry about plan should i follow, which diet, which book, which language, which guru to follow. But all of the "how" questions are really just a proxy for the discomfort of feeling dumb and confused and incompetent at something new. The only real answer is yeah, feeling dumb is part of it, just keep going.

Teaching is its own skill. Teaching well takes time to learn. Few expert programmers have training as educators.

In your experience, are mid-tier practitioners better at teaching? If so, that would be mean less support for my hypothesis.

But it's not true in my experience.

I learned this teaching my own daughter.

Teaching is a hard skill. You start with an explanation that you think works. Then you have to adjust it based on how your subject responds. Do they get it? Part? Is it not clicking? And then step out of yourself and figure out what's another way to visualize or think about the problem.

A lot of it is not about being right, but being useful.

Ive applied this to things you wouldn't think is teaching, like helping a friend understand why her career path isn't working out for her and steps she can take to improve it. Took an hour to figure out the bad assumption in her mind and figuring out a way around it.

At novice and intermediate competency levels you’re consciously considering what you’re doing as you plan it and do it. Conscious thought for most people is in the form of a semi-verbal “internal dialog” and is thus easier to verbalize to someone else. The downside of conscious thought is that you do it much more slowly and with much greater effort, but it’s easy to explain. For many subjects, once you’ve utterly mastered and internalized them, the way you think about them becomes largely intuitive, something you do in large part without conscious thought.

I lucked out and started working in a team where most people were there 5+ years and the real senior ones were 10+ years in the team. It was a cross functional team, somewhat isolated from the large organization, working on 'special projects' (government, military, police etc.) and it was comprised of engineers, testers, business analysists, technical project managers and a team lead. I spent so much time getting mentored, learning the ins and outs and being slowly ramped up to what I believe were great engineering practices.

I still attribute my engineering mind to that formative experience. At one point the company got bought out, teams dissolved and it became a hot mess where nothing was really done anymore. I left the team, with a heavy heart, and in a year's time the company was basically dead apart from some invested (crashed stocks so they must go up!) people that were left to maintain ongoing projects.

Since then, and it's been close to a decade, I haven't found any company where even a tiny sliver of hope exists to be able to mentor people the way I have been mentored. The landscape changed and everyone is looking for velocity. In my current role, whenever I ask a potential 'expert' if they have any experience with 'X' (due to their background in it), they just google, get to the first result and suggest that as a solution.

There's a very large gap of which the mid-tier is as natural to the expert as typing. I see this a lot in math, where the best lecturers are usually fairly junior.

My best friend has trouble reading graphs, as in y-axis, x-axis kind of thing. He doesn't know whether < means bigger or smaller. This is something a lot of experts have internalized decades ago, so we'd completely overlook how hard it looks to a beginner trying to learn comparison or AND/OR in a condition.

With respect, I think it is reasonable to assume post primary school education. The example given is perhaps a poor one.

He has a college degree, albeit non-STEM. This stuff is taught in middle/high school here if you're not in STEM.

Also I've taught coding bootcamps and after a few batches, I set up expectations that, no, you can not learn programming from "nothing" like the ads say. You're expected to know touch typing, basic math up to algebra, and how to open files on your own device. Those who don't know these things should get a refund and try again for the next batch.

Is it off-topic to simply say that expert-level explanations of some topics explained meaningfully may make for boring or out-of-reach lessons?

Is it useful to learn multiplication before addition? Or exponentiation before multiplication? Or tetration before exponentiation?

I mean there's arguments that if addition and multiplication both must be learned then maybe linear combination is a more useful end-state target for the lesson.

I'm not begging the question to refute your point. These are important questions, and I worry that society isn't seeking the answers, because as far as I know students still learn "times-tables" even though we have handheld computers capable of, from a photograph: reading, interpreting, and solving math problems from a description, so that those tasks are as contained as an arcane reference like integral lookup tables.

I see a world where students instead learn even finer machinery, such as logic, encapsulation, and organization rather than obsessing over whether they can rhythmically remember what 11x12 is for the rest of their lives.

Easy - they've forgotten what it's like to not know

I wrote about this 15+ years ago: https://antipaucity.com/2006/05/06/the-good-the-bad-and-the-...

Makes sense doesn't it though? When you learn everything you know without predefined structure, how easy is it to come up with one?

At the end of the day, self-taught like what they do - so if you really like what you do, you'll eventually figure out simply by spending time on the subject at hand...

e.g. Umberto Eco never replied to young authors who asked him "how can I publish my book?", simply because he did not had a good answer, he thought "same everyone publishes I suppose, you start small and then you go big and if the publisher thinks the book is good, they'll publish it"...

Expertise can't be taught. That's why you enter the workforce as a junior, whether you have an engineer's degree from MIT or you switched careers and are fresh out of a bootcamp. From there, the road to expertise is not paved with courses and books and hand-holding. It's paved with _lots_ of real world experiences, good and bad. Expertise _happened_ to those expert practitioners as it will happen to you as you fight fires, solve problems outside of your scope of responsibility, etc.

Probably a combination of forgetting what being a beginner is like and Berkson's paradox.

In my experience real experts are always good and teaching and synthesizing/summarizing and adapting to their audience.

Another matter is if an expert is approached by someone who does not put in even a bit of effort (looking for a mentor but rude, or did not do any prior research on their own).

Maybe it is too much of a tautology, but I would not consider someone that cannot communicate clearly an expert. The practitioner will need to put in the effort too, of course.

I'm by no mean an expert, but there is some degree of knowledge I might take for granted. When I write documentation that may ends up in from of a beginner, I ask for someone who doesn't in my field to read it and tell me if they understand all the steps involved, as I may be skipping a detail unintentionally because to me it is obvious.

In my experience, it's the gap in understanding. To someone with 10+ years experience, certain motes of knowledge might be long-worn givens, where to a beginner it could be groundbreaking wisdom. Without proper introspection, it can be difficult to correctly identify "What doesn't a beginner in this field know?"

In any field there could be a disconnect between the struggles experienced by the expert and the hand-holding expected by the beginner.

Programming is not just any field though. Programming is incredibly immature and filled with wildly unrealistic expectations. As a self taught developer of 25 years of experience I have largely given up teaching beginners how to easily write high performance code.

There is a lot of insecurity in programming putting tremendous downward pressure on bad habits. Teaching disciplined techniques to produce higher quality output is often met with hostility if contrary to a popular tool or current trend.

If the results of trying to teach produce such hostility why bother? Why would I volunteer my time to help people if the result is bitching, whining, and anger? Instead I will keep my experience to myself for my personal projects.

This reminds me of an exercise in 5th grade: write detailed instructions to be followed _exactly_ to make a peanut butter and jelly sandwich. The teacher would then follow the instructions exactly and hilarity ensues, teaching something about expectations, assumptions, instructions, etc. "I didn't know I had to tell them to _open_ the bread bag!"

It is a reason why I like to teach by asking questions so I can learn where they have gaps and help them see their own gaps. "How does the teacher actually access the bread?" "oh, we have to open the bag first!"

At best, if they remember and can judge their own education perfectly, an expert practitioner knows how to teach a beginner who is destined for expertness, which is a small fraction of beginners

Learning is a state transition from not knowing something to knowing it. Teaching is assisting someone through that state transition. To effectively teach someone, you need an understanding of their current state as well as the trajectory of states they are capable of transitioning through.

An expert is someone who knows the final state very well, but for whom the initial states and state transitions are a distant memory.

i've had experiences with high level experts in personal situations...a friend of mine likes to call them "one-armed crabs". They've worked so hard on making 1 hand incredibly strong (their specialty) - that a lot of other things fall by the wayside.

Those may include hygeine, grooming, diet, social skills, sensitivity to faux pas, extreme ineptitude in unrelated skills to their specialty (these can get to absurd levels) and so on.

So it stands to reason perhaps that their ability to illuminate is next to impossible.

The road to understanding can be complex, filled with head banging on the wall...if you think to yourself what exactly made you go from "i dont get it", to "Eureka" - a lot of times there is no answer. Sometimes the synapses just do their thing. Now - previously mystic concept is intuitive, how do you go about translating that?

The curse of competence - Once you’re unconsciously competent at a task then many / all sub tasks become “obvious”, since they’re all unconscious know they’re difficult to enumerate to a beginner.

I’ve never met a real expert who couldn’t teach. Sounds like you haven’t met real experts.

https://commoncog.com/tacit-knowledge-is-a-real-thing/ it's been posted on HN a while ago.

tl;dr; Tacit knowledge is knowledge that cannot be captured through words alone. Think riding a bicycle.

Sometimes the beginners are going to need a much longer attention span . . .

Other times it's the experts . . .

Reply on news.ycombinator.com