Hacker News new | past | comments | ask | show | jobs | submit login
How to Make Other Developers Hate to Work with You (anaxi.com)
272 points by turingbook on Feb 22, 2019 | hide | past | favorite | 179 comments

I thunk there's a second "flavor" of arrogance, where a person believes they're the smartest, but not most knowledgeable, in the room. These people will openly admit they can improve and seek training, partly because they believe they're capable of anything. They don't think they're at the top of the mountain, just that their mountain has no peak.

The real problem is they believe "lazy" coworkers will eventually be below them in every subject. They may seek out colleagues' advice some times, but likely will never return because they "get it" now."

They'll act like a "jack and master of all trades." This can be nice in small teams, but usually leads to components nobody (not even the arrogant one) can understand let alone maintain. So experts need to be brought in, anyway.

These people let the quick ascent of "Mount Ignorance" multiple subjects get to their head. No such thing as a Renaissance man anymore.

Sadly, I'm that flavor of arrogance.

Arrogance has very little to do with thinking you know a lot, or disagreeing with others, and everything to do with failing to follow social protocol. If you let other people speak, pause long enough that they feel like you are giving their words a fair evaluation, and then don't use words that they must interpret as telling them they are wrong you will not be seen as arrogant. This is true even if you rarely or never agree with anyone.

- let people finish their statement: If you don't let someone finish and you interrupt, they will decide that you are just refusing to hear them out rather than disagreeing. It doesn't matter if you can prove mathematically with extensive sources and footnotes, most of the time they won't be able to accept that you understood their point and showed it to be flawed.

- pause long enough after they speak: Again, if you instantly explain why they are wrong, most of the time other people will not be able to handle this. They will decide you didn't actually consider what they said. It doesn't matter what is actually true, and if you somehow do convince them that you are right they will be doubly resistant since you are making them feel stupid on top of everything else. A huge benefit of this and the previous bullet is that you will often find that at first you misinterpreted what the person was saying, so you were about to disagree for no reason.

- don't force them to admit to themselves that they were wrong: if you let people 'save face' they can just get on board and agree with you. If you force them to grapple with being wrong they will be difficult and will hate your guts. It may seem silly, but replying 'I don't think that will work because ...' is going to be the source of turmoil, when saying 'Interesting. Previously we did something similar to that but we had a problem where ...' is going to win the other person over instantly most of the time. Later they will see that they were wrong, but you didn't rub their nose in it and force them to grapple with it in a social setting where they are going to have to feel shame.

Failure to do these things is just what people mean whenever they say someone is arrogant.

It has taken me a few years and I still fail sometimes, but I have learned to let people finish and then ask questions instead of telling them they're wrong, even when it's totally obvious to me. "What if that fails in this way?" instead of "that will fail in this way!". It makes the discussion collaborative instead of assuming that it's my job to point out everyone's failures. The punch line in one of my favorite comic stips is "It's not healthy to keep your problems bottled up inside me." It took me a long time to see how that applied to my own interactions.

The thing that gets me is the same thing I hate about politics. You build an elaborate plan on a bad supposition, and people hear the whole plan and forget it’s based on nonsense.

If the ground premise is wrong, and you see where they are going with this silliness, it’s challenging on multiple levels to “let them finish”.

It's mostly about letting them finish the sentence, rather than the plan.

Point 3 though. At some point how can you be so demure? Sometimes people are just stupid and wrong and need to get checked. Some guy working for a year getting nothing done and pretending like its some other problem than what it actually is needs to be forcefully briefed on their ignorance not 'hrmm well you see we tried it this way'.

> needs to be forcefully briefed

This will let you feel superior and get the pleasure of making the dumb person feel terrible, but it really won't accomplish anything productive. What do you hope to gain by doing this? Will they start being smarter? Will they be better at their job?

I'm sorry, doing what you describe is just being a bully. If management isn't dealing with the problem person you should privately make them aware of the issues (in writing if possible). If management is aware of the issues and won't do anything about it, well you did your part and you aren't allowed to fix the problem. Unless it is a life and death or national security situation, either accept that the dummy is going to be around or resign.

If they haven't done any work for a year their manager is useless. You work at a place where managers don't do their job. If you take it upon yourself to set this person straight, you are basically propping up this useless manager. Useless managers hire useless people and chase away good people. Fixing this problem treats the symptom and may let the bad manager last longer than they would have otherwise, which means you are going to have to go set two additional people straight about how useless they are next year.

I agree it's a management problem. However, I've encountered the case where a developer abruptly teleports from "plateau of sustainability" (high confidence, knowledge and experience) instantly to "Mt. Stupid". And every time it was a case of users becoming confused with UI and doing something that caused injury (including data loss). In that case the developer consistently blamed both the user and user advocate.

I do not think it's bullying to give a bully a taste of their own medicine. It may not solve the problem, to be sure, but pushing back on bullies is not wrong. Pointing out their cognitive dissonance, and drawing a line from their stubbornness to being a bad developer is not wrong either. In particular where the manager doesn't care because 90% of the time that developer is delivering the goods, and just doesn't properly assess the ensuing (UI/UX) hostility.

Bad UI will piss off users. It's I completely expect they should express that frustration directly to the responsible developer. If that developer can't handle it, both accepting the criticism and making appropriate design adjustments, get out of the UI/UX development business.

This is good advice.

I find situations where I feel someone else "needs to be forcefully briefed on their ignorance" are situations where I likely don't have the rank, influence, or audience to improve the situation anyway.

I don't really subscribe to Kant-style categorical imperative thinking or defeatist attitudes, but in this case I've always made better progress changing an existing bad thing by being "demure" and accommodating. I might have agency to change things, but if an group of others have have agency to keep it the same, failing to cooperate is counterproductive unless I entirely opt-out or leave.

If it's actually that bad, you should just fire them or do your best to not work with them because they will drag you down with them.

For milder/more moderate cases, giving someone a way to concede without publicly admitting they're wrong is a often a useful tactic to get said person to do things the right way while building trust in your relationship with them.

> Some guy working for a year getting nothing done and pretending like its some other problem than what it actually is

This is a management issue. In a rational engineering organization this guy's manager would be like "okay, so we know there's a problem with this tool. Now how do we work around the tool so we can move on to another problem?" And if the conversation becomes "workarounds are stupid we need to make them fix their tool" then the manager needs to be all like "I agree with you in a perfect world that's how things would work, but we're engineers being paid to solve people's problems and not to build the perfect software system. So build the workaround so we can solve the problem and we'll circle around and fix the tool when we have time."

Sounds like they have a bad manager. Yes some people are bad at their job, and that includes developers, but in every case I've encountered it, their manager enabled it. It could be an equally arrogant bad manager, or a milquetoast bad manager. Servility in a manager is a bad attribute. They should be someone who seeks out the demure contrarian and acts as their advocate, should the manager find their position compelling. Not everyone is confrontational, that isn't the problem though. If the manager isn't seeking out contrary opinions, and only acts obsequious to their developers? That's a stool pigeon of a manager. Bad manager and bad developer? Good luck with that combination.

I kind of miss the point where you think about what people said, react to that actual argument and wonder for brief time whether they actually might have a point.

It is failure of this one way more then just lack of pause - especially in long term relationships.

> don't force them to admit to themselves that they were wrong: if you let people 'save face' they can just get on board and agree with you. If you force them to grapple with being wrong they will be difficult and will hate your guts

I think that an inability to admit a mistake is arrogant and results in a lack of accountability and ownership over individual's work. If someone makes a mistake they should be proactively owning it and trying to find solutions to avoid similar mistakes in the future.


Please don't cross into personal attack like this. Comments here need to be civil and substantive.


Your claim was:

> if you let people 'save face' they can just get on board and agree with you. If you force them to grapple with being wrong they will be difficult and will hate your guts

Your response to my post:

> I doubt you properly understand that other people are full thinking beings equal to yourself since you assume everyone secretly agrees with you about every detail of everything all the time and the only issue is whether they are going to admit it or lie.

To me, you're contradicting your own position. It also seems to me that you're the one that doesn't "properly understand that other people are full thinking beings equal to yourself" based on the arrogance and condescension of your response.

Can you unpack exactly what made this comment go from 0 to 100? I didn't get any of that from the parent post.

That take might be overly cynical. People who have a jack-of-all-trades skill set might not necessarily build the best components themselves, but might be able to bring some big-picture insight that more specialized developers don't have the awareness for, leading the team in the right direction.

I'm not suggesting you transition into becoming a PM or Architect or anything like that, just saying that developers like you count, too. If you have a decent work ethic, then you and your unique brand of knowledge-seeking have something to bring to the table.

> "usually leads to components nobody can understand let alone maintain"

As you advance up the mountain, you will learn to see complexity simply. When you make this realization, your components will simplify dramatically. You must learn to see simply before you can design simply. It just takes time. There's nothing wrong with this, humility doesn't mean giving up your belief that you're capable of anything.

We were all born without any knowledge. We all learned everything we needed to know to get where we are today. We've put men on the moon. We've sent probes to escape our solar system. We've sent submarines to the bottom of the ocean. Humans have the capability to do anything we put our minds to. You can too. Everything you understand today, you've learned in the short lifespan you've had. Today you will learn more, and tomorrow and the next. Just because you find humility in your place, doesn't mean you're not capable of anything.

As individuals, there's damn little we can achieve.

Nobody said you must do it alone.

> doesn't mean giving up your belief that you're capable of anything.

> Humans have the capability to do anything we put our minds to. You can too.

> doesn't mean you're not capable of anything.

That does sound like you're talking to individuals. Turn down the hyperbole a notch and I might even agree

I have been concerned about occupying a similar space myself. The great secret that we all do well to remember is that being humble is simply a matter of practicing being humble. In particular, when I treat other team members as experts in what they do, and trust that they'll value me for whatever it is that I do, I become much more efficient, and feel more relaxed too. Yes, there are still times that I notice myself acting vain, but these days a few slow, grounding breaths goes a long way toward resetting me.

In fact, I wonder if the narcissism you speak of might have more to do with insecurity, deep down. In my experience, when I see myself taking a behavior that I consciously am against, the cause tends to be emotional and the solution tends to be to gently, curiously observe those situations and start questioning how I feel. When I love and take care of my emotional world, it becomes way easier to interact in a way I feel is more optimal on the fly.

I just read the article and was like “I don’t think I suffer from these much at all” then read this comment. This is very insightful and also something I just realized that I greatly suffer from. Thanks.

I relate to this.

I think one of the symptoms is not picking up enough detailed understanding of anything because you keep moving around and don't believe in most external knowledge anyway (eg. don't want to learn [beyond a certain depth] SQL, it's so ugly on some inner level I'd much rather start with datalog and build my own up, don't want to learn economics I don't trust those science cargo-culters). But the other reason you don't want to specialize is that it would be to accept a limit on yourself (OK I'm going to get specialize on getting very good at c++, my upper salary cap is 500k and likely cap 200k, career is mapped out).

One of the things that's slowly curing me I think is running into examples of systems that are genuinely very heavy to understand/improve, and examples of people who are just unarguably higher ceiling in different ways than you.

eg. go watch a SGM play 30 second hyperbullet chess, I don't think I have either the memory or raw processing speed to do what they do, ever. If those guys exist a lot of other people have way higher ceilings. Also realize that chess (and by extension must be a lot of other activities) rely on both a crazy cpu, an amazing memory and an insanely built-out internal database of positions and ideas, and since my cpu is good but not unbelievable, my memory is nothing special and my time is limited there must be a lot of activities I'll never really be good at. Also just getting older and staying unsuccessful relative to your ego, you just start to compromise I think.

> SQL, it's so ugly on some inner level I'd much rather start with datalog and build my own up

SQL syntax may be ugly, but for once, it's an example of a widely used technology that is based on a very firm and well thought out theoretical framework.

As a developer, it is definitely worth investing time in understanding relational databases and the theory behind them.


Thanks, I've read arguments both ways and am trying to study logic/logic programming.


Nobody’s mountain has a peak unless you either stop learning or die.

Everyone's mountain has a peak. We have a finite amount of compute power in our heads, and a finite amount of time on the earth. That means we have a fixed amount of complexity we can work through.

That said, fixing someone's shitty microservice code is probably not bottlenecked by our cognitive capacities. It's probably bottlenecked by financial constraints.

At the point where all your practicing is just maintaining your performance, you've reached your peak. You can stay there, or you can go higher if you practice more (or better). But there is a finite amount of practicing you can do each day. (Studies argue that 4 hours is plenty.) So there is an absolute limit per individual.

Are we talking about a physical skill or a personal capacity for knowledge? Because while your argument might have merit when applied to something like sports, it doesn't hold much water when talking about accumulated knowledge. Yes, I might forget how to do a math problem the way I could when studying for a test in college, but I'll never forget that knowledge exists and where to find it if I should need to relearn. The metaphorical peak I'm speaking of relates to a person's ability to index knowledge like a search engine for the internet.

>Yes, I might forget how to do a math problem the way I could when studying for a test in college, but I'll never forget that knowledge exists and where to find it if I should need to relearn. The metaphorical peak I'm speaking of relates to a person's ability to index knowledge like a search engine for the internet.

The answer to your problem is actually and always "the internet". There, you've peaked.

Well at least it was after high school

Mental performance needs practice just like physical performance. At some point you max out. And if you don't keep up practicing at level, your performance declines. Knowledge too is bound by those rules. We're constantly forgetting as we learn new stuff.

> I'll never forget that knowledge exists and where to find it if I should need to relearn.

Meaning you'd have to practice to again reach your former performance. Sure you may reach your peak level again and reach it quicker thanks to your old practice. But you do need to practice again. And then again, if you want to keep it.

True, but I just need to stop thinking I can climb every mountain faster than people focusing on a couple. To put it another way, I need to rely more on others' progress. Even better, help them progress.

Yes, but different people are climbing different slopes; everyone can keep climbing, but that doesn't mean they'll reach the same height.

Yes, but those slopes are indeterminate from an outside observer because success comes from within. For every person you think you might have pigeon-holed, I'll show you another that broke the mold. You see everyone starts off facing a metaphorical vertical cliff in anything they attempt. It's over time that you learn the passages and tools that make it easier to climb; that lessen the slope. Unless we're talking about physically dunking a basketball, anyone can be a savant in something with enough effort. But again that future effort is not knowable to a 3rd party, they can only base opinions on past observations. People can change and they can change in an instant.

Thank you for your admission. I too am naturally arrogant. People think I'm being weird if I'm not arrogant. I get weird people saying they don't believe me when I'm being humble. Probably because arrogant is who I am as a person. I feel people like me more when I'm being my arrogant self. Oddly, when I was my least arrogant and most humble I exhibited most of the traits in this article.

I was quite sour over the fact that I felt like during the interview process I was lied to about the position and what it would entail. Naturally, I was a bit salty about it and the fact that people were doing things that I would get criticized for. It ultimately hurt my confidence in myself. This made me exhibit most if not all of these traits during my stay at this company. I was trying to make lemons into lemonade. Ultimately I was able help us get on the right path and I was happy with my contributions to the team even if it did ruffle feathers and those people hated me afterwards. I took their criticism and I still try to use it to be a better person to work with.

This take can cause a pretty unproductive worldview. The trouble is, we've all met different people who fall into broad categories like "jack of all trades". A lot of morons will say this sort of thing. Sometimes it's because they think you think they are dumb, so they want to show you that they are good at other stuff so as to keep their ego happy.

However, generalists are a very useful component of a team. They tend to be able to have a higher level view of all the moving parts, and typically are great for, say, proposing integrations, or high level system design. Also, often specialists will solve a problem in their specialization, fit be damned. They need someone to derail them and say e.g. "this would be much, much easier if we leaned on MySQL to do this".

So, this feels like a baby and bathwater situation to me.

I'm sorry to read so deeply into your anecdote, but there's some interesting stuff to unpack.

> This can be nice in small teams, but usually leads to components nobody (not even the arrogant one) can understand

This describes nearly every company's MVP, so if you're thinking of a person in particular who had to cobble together some dynamic mess, chances are they have great regret for their design. But remember, tech debt is a luxury in these cases - often the alternative is no job or company, as the runway was exhausted, or investors were not impressed.

> A person believes they're the smartest, but not most knowledgeable, in the room.

I think there's no getting around this. In fact, TBH I hope my colleagues feel this way - I want them to advocate for things they think are better, and I don't want them to instead fall prey to their own insecurities.

It's actually kind of a little self-help bullshitty -- the "you can do anything" stuff -- but truly it's actually not awful advice. I guarantee there's someone you work with who has a better way of doing things, but isn't confident enough to endorse it.

Anyhow... thanks for the writing prompt? ;)

I'm the worst for this. I firmly believe that given a totally novel problem in any domain that neither of us are familiar with. I will always solve it before you.

I also believe that whatever domain you're currently in I can outperform you in 5 years MAX.

It's crazy the bullshit we believe that 'feels' right day by day

I see this in my step mother. My family isn't very high in social standing, but she has a nursing degree and a long well paying career, better than anyone else in my family. Now that I have a CS degree with similar pay, she pays me respect but still tangentially riffs on things she only half understands.

Just don't settle for being a 'jack', become an expert, and produce something you think is maintainable for yourself and your colleagues.

> Just don't settle for being a 'jack', become an expert

Expert in what? There will always be people "more expert" than me in any particular domain (specific tech, hobbies, whatever). At some point, there's diminishing returns in becoming 'expert' in things you don't need to be an expert in, when "competence" is enough.

In many offices, mere competency makes you the default expert.

Sure. Was just pushing back on the "don't settle for jack-of-all-trades - become the expert". Well... that's a never-ending goal. Perhaps I want to be an expert jack-of-all-trades?

I can't be the expert in all things, or even most things I want to be. There's simply not enough time - people with 10 years of experience on me will always be 10 years ahead - and usually that means there's more 'expert' than me on XYZ (not always, time != expertise exactly, but it's often an signal).

Enough to produce something maintainable, that you can be proud of

The problem with becoming an expert is that you greatly limit your marketability and pigeonhole yourself into a small niche. When that niche loses demand, or you want to move someplace where there's no jobs in that niche, you're screwed. A generalist has a much easier time changing jobs and therefore more job security.

The whole point of the arrogance / humility axis is the awareness that there is not a destination to be arrived at named "expert". The DK graph goes up and to the right, to infinity and beyond. We must merely try to do our best, and try again.

If you think it's a problem that you have this type of arrogance, what are you doing to improve and seek training about it?

I keep in mind that I tend toward arrogance and think twice whenever "oh, I can do that" pops in my head. Also, splitting tasks with a team means other people can grab what they do well and I won't overreach.

9. Refuse to do any maintenance and only work on new projects.

10. Hide your actual productivity. IE: get done with your coding tasks in half the time, leaving other devs have to fix all the bugs that QA kick back. (This is probably a managerial issue, actually, but perpetuated by devs who do # 9.)

#9 for sure. Probably a side effect of being sloppy or writing bad code.

My version of #9 I'm dealing with is: Is this an feature that a customer will use? Or is it internal improvement? Yes to first, no to latter.

Are you saying that your team hasn’t done enough of the latter, or that they don’t focus enough on the former?

We haven’t done enough of the latter. I’d love to automate and improve a lot of stuff but our project manager is pretty much always client visible first.

#9 might be a symptom of a structural problem. If developers routinely encounter problems arising from technical debt, and are continually asked to fix those problems, but are somehow never afforded the opportunity to retire the technical debt, that creates an aversion to the maintenance work.

"If it ain't broke, don't fix it; if it is broke, just use duct tape."

A lot of places have infinite budget for fixing things an a patchwork fashion, yet zero budget for planning ahead for the maintenance phase, or for scheduling preventative maintenance. It's the equivalent of never going to the doctor for a checkup, and only going to the Emergency Department at the nearest hospital when an organ explodes. Nobody wants to be in a long-term relationship with an exploding-organ person, and it's the same for code.


#10 is an organizational issue. If a dev has "ownership" of a particular domain, they will naturally work very hard in the short run at setting it up so that they may be as lazy as possible in the long run. But if they are continually kicked off their catbird seat of smoothly-running machinery in order to clean up other people's messes, then there is some incentive to exaggerate the effort it takes to clean up their own mess. Or to make it appear as though they have a mess to clean up.

Even worse, if no one has ownership over anything, there is incentive to always drag heels on the current thankless and tedious task just to delay assignment of the next thankless and tedious task. Do the minimum amount of work to fix the current issue, and let the next guy deal with the fallout, even if the next guy is also you. Because if you bust your ass to fix it right, the next guy won't thank you, because there is no next guy. You can only work on the stuff that is broken, and if there's not enough work to do, people get laid off.

And that is the equivalent of under-staffing the POS cashiers, so that none of them are ever idle. That strategy results in fewer customer transactions, because when cashiers are never idle, that also means that there is always a wait for customers to check out, no matter what size the order, and the cashiers have zero individual incentive to complete an order quickly. Monitoring scan rates only covers the problem, which is that employees are implicitly punished for doing their job too well. Go to Wal-Mart, and all the cashiers are always busy, but also always dragging ass, and all the check-out lines are three customers deep. Go to Publix, and cashiers with empty lanes cheerfully offer to check you out the instant you have found everything you need, and will direct you to the correct aisle if you haven't. Idleness is not just doing nothing, but is also customer responsiveness.

When other devs do this, your best tactic is probably just to follow their lead, and possibly collude with each another against management for your own benefit. The org is broken, and you won't be allowed to fix it.

> #9 might be a symptom of a structural problem. If developers routinely encounter problems arising from technical debt, and are continually asked to fix those problems, but are somehow never afforded the opportunity to retire the technical debt, that creates an aversion to the maintenance work.

I've seen a couple fresh grads come in with that attitude and get away with it. I think it is more personality type than anything. What I saw usually happen was they were allowed to work on a new feature as kind of a treat for joining, but after when tasked to start fixing issues from legacy, they flat out refused. Even taking it to the head of engineering and making a stink.

There is a sweet spot.

The boss can't hire new grads to work on it, because they can smell the manure from a mile away. So as long as you're willing to slap on yet another coat of superficial patches, they'll keep paying you to do it.

That's an equally-bad symptom of the same problem, approaching from the opposite side. The structural problem still doesn't get fixed, because the developers aren't repulsed by it at a visceral level quite enough to kill it with fire or nuke it from orbit. It's just scary enough to scare off the competition.

The technical debt is kept in place, so that people can live off of the technical interest payments. It's not an honorable living, but it is a living.

You're still stuck on "the code is bad and that's why they don't want to touch it." I'm saying, their attitude is bad and feel maintenance is beneath them.

Every, and I mean every, development shop has some amount of tech debt. But even if the code is generally good quality, you have a subset of devs who think maintenance is below them. That's a bad attitude and people don't want to work with that.

It's worse than that. All code instantly becomes technical debt the moment it's deployed. A development shop is almost entirely composed of technical debt, unless it never shipped anything.

Or, augmenting #9, the team lead who takes all the interesting work and doles out compliance etc to everyone else.

I really don't know how a team lead would find the time with all that turnover. Training is a bitch.

Sorry, but real life situation is infinitely more complicated.

Crybullies, manipulative people, people driven by greed to have power over others, hazing the new team member for whatever reason. Some are much more likely and zealous to report colleagues to the supervisors, some less. Some are more likely to be listened by the supervisors, some less.

In unlucky circumstances, couple of rounds of conversations with managers and suddenly you are the asshole nobody wants to work with.

> Developers can’t easily go back to where they were right

> before an interruption. They need to get into the mindset

> for development and then slowly trace back to where they

> left off. And every fellow developer knows that.

I used to think like that too, but after 10 years programming in many environments, now I'm inclined to disagree. If I cannot refocus immediately after interruption - I know I'm doing something wrong. Maybe the thing is too complex and must be cut down to smaller pieces, maybe I didn't keep enough notes or thinking through - anyhow nowadays I cannot work without a clear understanding of what simple piece of what complex piece I'm doing and where I've written down the other pieces I should do next.

Being able to reliably refocus after interruptions is an enviable skill.

I've learned some of that (especially when I was an intern divided 50/50 between development and interrupt-driven sys-admin :), but for me, interrupt cost seems to vary widely with the demands of the task and the nature interruption.

Web/GUI front-end work is usually very interruptible for me, for example. Usually, most of the time, I have in mind "I want to get to roughly this, and I just need to spend time getting or making the tools get there", and it's often easy to see where I left off. Coding seems usually mostly proceeding along a series of small tasks ("those controls are done, and next is that control"), or iterative refinement ("these two elements could be combined", or "this part of the layout is poor when rotated"), which are often very visible, and easy to get into.

However, if, for example, I'm in the middle of designing an algorithm, and trying to cover every case in a good form, and trying see opportunities, and I've mentally "paged in" a lot of domain-specific information and scenarios and such... an interruption at that moment with something that requires a very different domain to be paged in, and substantial thought... could really ruin the roll I was on with the algorithm. Even if I have a tentative skeleton with quick comments sprinkled through it, for the current thinking and known questions/opportunities. I'm simply not smart enough that I can't sometimes lose something with a big interruption at the wrong moment.

There's also the theory and meme of "flow" [1], but I don't understand it well enough to say whether it covers the impressions above.

[1] https://en.wikipedia.org/wiki/Flow_(psychology)

To me that just sounds like you don't have tasks that require deep thought. And that you've developed habits to cope with interruptions, such as note-taking.

To turn a complex thing into a list of simple things can be done by experience as well as by deep thought. But don't think you could get that experience in 5-minute chunks of thinking. To get the experience you need deep thought.

Maybe set your sights higher if it's all easy to you? Or enjoy the fruits of experience, of course.

To me that just sounds like you don't have tasks that require deep thought.

If everything you do requires deep thought, this is possibly a sign of bad code design.

Not _everything_ but a large part of what I do does involve rather deep thought.

But indeed, what helps me is to pick one of the things that is easy to 'refocus'.

Not _everything_ but a large part of what I do does involve rather deep thought.

Anyone's ability to think deeply and pull out feats of clever is a limited resource. A well designed project tries to optimize this. Think of it as martial arts. It's good to be strong. The techniques are there to use your strength more efficiently.

I was once in an Aikido class with someone who lifted. He had to constantly remind himself to use the technique and not just strength. (As Ramsey Dewey likes to point out, technique plus strength is pretty darn awesome.)

Haha, could be yes. And at the opposite end:

If everything I do can be achieved without thinking hard I'll be replaced with a tiny shell script written by someone who does.

If I cannot refocus immediately after interruption - I know I'm doing something wrong.

This can be a sign that code needs to be reorganized. I've seen a number of really good coders in their 20's who are admirable for brilliant insights. However, they fall into this pattern of thinking of themselves as invincible, and they take on tasks where they are doing the "heavy lifting" of remembering tons of details all at once.

If you can lift 400 X's by hand, then you could easily lift 4000 X's with the right tool.

One of my bosses kept a stack, explicitly, and used that to keep track of his excursions into sub and side tasks.

I used to say that being a good programmer was being a professional forgetter as much as a professional memorizer. Basically, you should be constantly reorganizing your code as you go, so that there's no danger that forgetting a detail will result in an error. In a way, it's like people who work on high platforms who take precautions so that one drop or one slip won't result in injury or fatality.

In other words, work smarter, not harder.

> Maybe the thing is too complex and must be cut down to smaller pieces

i suspect that for many, it's not necessarily their fault Thing is 'too complex' (though perhaps this is tangential); nor are they necessarily provided the opportunity, time or even permission to 'cut down to smaller pieces'. piles of work, looming deadlines and all that.

I've found that the issue of re-acclimating to context has more to do with the quality of my notes than with the complexity/simplicity of the question.

In the course of figuring out something complicated, it's sort of like a tree traversal. If I take good notes on the leaf nodes I need to fill out, then it's easy to just resume right at the leaf node. If I don't take good notes, I have to retrace the rest of the tree.

One of the things I do to offset possible interruptions is to write things out or visualize them in some way. It helps me pick back up on the track I was interrupted from, but also has the added benefit of helping visualize the problem, which often leads to a greater understanding. It's a little bit of overhead that can pay off.

I agree, right now I find it hmm weak excuse. If someone is junior then OK, but if someone is claiming 5yrs of experience I am not getting those kind of excuses from that person.

Here's a profile of a developer:

- Easy going and pleasant.

- Really interested in programming and tech in general

- Talks intelligently about the problem at hand. Asks all the right questions. Agrees a plan of action in collaboration with colleagues.

- Creates systems that definitely appear to work

- Code looks pretty sane. Structure is right. Style seems good.

- Always very responsive to problems. Jumps straight on them.

- Every problem fixed another one seems to pop up.

- Eventually you dig into the code and problem space in detail and you find that in a very subtle, even complex way the whole thing is completely wrong, the problem space misunderstood and everything patched over with little fixes to get things working for a particular case.

In the end no system created by the person is useable without major remediation by another developer.

Anyone experienced this?

I have definitely experienced this.

Imagine a competent developer writes a mostly correct system with few bugs, and that requires few fixes over time.

Management will likely perceive the first incompetent developer as better than the more competent one.

Why? Because the incompetent developer is seen to be fixing problems all the time, which is a positive thing.[1] (You could say his work is more 'salient' to management.)

Frustratingly, the incompetent guy comes off as a rock star.

[1] Ignorant of the fact that the problems were created by the developer himself.

In big companies maybe, in small company I would say I took over job of guys that were firefighting awesomely.

If they would continue doing that, company would go under. When I joined, big customer was already starting to pull out because there were constant issues, though fixed by brilliance of those guys.

With me on board and those guys gone there is a lot less issues and guess what... My boss has time to drive around country to bring in new customers and we are growing.

When there were issues my boss had to stay around in office to keep track of problems and fight hard to keep customers from leaving.

I don't say that your story is not true, but those guys in your management are lamentable, they don't understand how much more could be done if they would not have those problems in first place...

> I don't say that your story is not true, but those guys in your management are lamentable, they don't understand how much more could be done if they would not have those problems in first place...

In large companies with more layers between developers and the customer, this kind of dynamic can go on for a while.

I've seen it many times.

> - Code looks pretty sane. Structure is right. Style seems good.

You're attributing that bullet point to the wrong developer.

Here's a profile of a reviewer:

- Isn't paying sufficient attention to the code being submitted.

Don't get me wrong-- I've ripped away drywall to reveal nastiness before. But what you've described is systemic, and for that the reviewer must share some responsibility.

Look I'm happy to take my part of the blame but some of what was wrong was so damn subtle.

Ultimately one does trust people to make design decisions at some level of code that one does not second guess...

But you wrote that none of the systems this developer implemented were even usable without intervention. What's the point of reviewing code if you don't eventually read some of it in enough detail to clamp down?

Or at least revisiting the design of your tests.

Not before 5 or 10 years.

Patches and fixes have accumulated in response to strange edge case and bugs you wouldn't believe if told.

The organization and the problem space have shifted somewhat over time. A few things could have been done differently in the new environment.

Sounds like insufficient testing to me. This kind of deep/far reaching issue should have been crashing tests written before others would have seen it.

Assuming that there is some testing in their work (if not; there's the answer), my guess would be that it's largely sanity testing of each functionality in a vaccuum. The coaching would probably be to demonstrate more integration-style testing, possibly using whatever tests caught system-level issues that the dev has pushed in the past.

Could also be organizational. Constantly shifting whims of stakeholders, etc. Yes, requirements change but changing major requirements without a major refactor can cause these kinds of problems with the code. Or if the developer is being given unreasonable deadlines, no time/resources for testing, etc etc.

>the problem space misunderstood and everything patched over with little fixes to get things working for a particular case.

Sounds like testing found issues which were hacked over, but more will keep popping up because the design is inherently incorrect.

No, it's just lack of true skill.

Yes. Very enthusiastic, outspoken, easy going, pleasant to work with, but really can't complete the tasks without someone going back and fixing it.

> easy going, pleasant to work with

This is a survival tactic, and in a way it's a dishonest put on.

If the same developer were abrasive arrogant and difficult to get along with, then the other programmers, who always have to fix his work, would have him run out of the company.

Meanwhile, sometimes the crustiest curmudgeons in a team are the ones who have the most experience and knowledge to impart if you get to know them.

Because they're fixing some other problem they want to work on. Not the problem they actually have. They know it.

> A team with a bad developer is way better off short one developer than it is with a bad element.

How do you know that you just didn't communicate your issues well to the person? Is there a way to help them, you and your relationship improve so that you can be a stronger team?

We all have blindspots and weaknesses, and weaknesses that fit together like puzzle-pieces tend to find each other until you resolve them.

The half-moon is only an illusion, the whole moon is still there. You just don't see it, and you need the whole.

> How do you know that you just didn't communicate your issues well to the person?

When you encounter that person, you'll know. Some people just aren't capable of grocking certain things.

I had a guy on my team like this. He just wasn't getting it. At first, I would spend extra time with him, communicating the issue from every possible angle. He would almost always end up producing something so far off base. Everyone else on the team was sync'd but he just never got it.

The best part is that he refused to believe that he was incompetent. I say "best", because that gave me an easy way to get him to move on without firing him. I knew if I wrote him up one time, his pride would kick in, and he would start looking for another job. The plan worked perfectly.

I think #1 and #6 go hand in hand, and usually results in #2. If you disregard your team, and you're arrogant, your code will likely be sloppy as hell. In my opinion that makes you the worst member of the team.

I spoke with a coworker about this and we agreed that you just need to be humble as a software developer: we all are going to screw up, we don't all know all the answers, and we don't know why others in our teams say / know specific things, but we should not take their advise lightly it might of been a painful lesson for them to begin with!

My girlfriend being a developer also we end up having this kind of conversation very often and we agreed to say the developers we enjoy the most to work with are knowledgeable but yet humble persons. In my opinions personality traits are often disregarded over technical skills during interviews.

The best places to work for try to balance both.

Holy crap, is

>gives cryptic names for variables, or at best not self-explanatory

sure a big one. It's part of a category of small decisions that dramatically increase cognitive load when accumulated. Along with it:

>Passing many arguments into functions with names that have no relevance to their role inside the function, so you constantly have to look up which argument maps to which function param

>Needlessly reorganizing data structures or renaming variables so that it becomes more difficult to reason about the flow of data

Naming things isn't just the hardest problem in programming, it's also the quickest way to piss off your peers if you're lazy about it.

My coworker used to be a solo developer and would make up recursive acronyms for every variable name, like:

  MSA: MSA subject analysis
It's pretty hard to go through scripts older than a year and understand what's going on.

A former primary developer of my code base had something wired backwards in his brain. The code is full of double negatives:

if(!FeatureNotDisabled){ .... }

It drives me absolutely bonkers!

For me this often happens incrementally. Like they started with "if (FeatureNotDisabled)" and then locally refactored the code until it was "if (!FeatureNotDisabled)". The problem is that they then didn't go back for another pass before commit, and reverse the double-negative by changing the variable name and semantics. This could be due to laziness, ignorance (it works, what's the problem?), fear (it works now, let's not break it), or time mismanagement (it took long enough to get it working, no time to make everything "perfect").

I could understand that evolution, but I have become convinced that this particular developer simply worked with negative logic in his brain. Even the configuration files will have:




Instead of using feature flags to enable a feature, he used feature flags to disable features. He was very consistent and the logic is all correct, simply inverted.

That's a triple negative! (!, Not, Disabled).

Yes, `if (!FeatureEnabled){ ... }` would be clearer.

> That's a triple negative! (!, Not, Disabled).

It's only a double negative (“disabled” is affirmative, though the same sense can be communicated by the single negative “not enabled”.)

Another one is giving insanely long names but only changing one little part, instead of taking advantage of the heirarchical nature of programming languages and coming up with a nice nested structure. Example:

  class TheRpcThing
    object TheRpcThingHost
    object TheRpcThingHostTimer
    class TheRpcThingValidationException
      object TheRpcThingHostValidationException
      object TheRpcThingHostValidationExceptionErrorMessage
      object TheRpcThingHostValidationExceptionErrorMessageFormatter
The cognitive cost of reading a bunch of previously-unseen code, which has an accumulation (the perfect word for this kind of thing) of the above can be so frustrating.

I actually use the number of capital letters in camel-cased variable/method names as a signal that the code might need refactoring.

Generally, every capital letter indicates another hierarchical relationship from left-to-right. Usually these are implementation details that shouldn't be publicly exposed, or at least could be separated out/encapsulated better in a different object/function structure.

Obviously it's a rule-of-thumb and there are exceptions, but I've found this works pretty well.

Gah, naming! It feels like a sign of a lack of empathy to not at least give some semi-meaningful name to a variable. "x" is only acceptable if you're literally writing a function to perform algebra.

There's not a single thing in there that I don't recognize in myself at one time or another.

I sometimes feel like I don't see anti-patterns as I go through my career, I fall over them and sprawl on the floor.

The important thing is that you can recognize your own problematic behavior, and actively work on fixing it. Nobody's perfect, but I imagine there's a reason the author put #1 as #1 ;)

Yup, looking at common problematic behavior without reflecting inward, is an issue in and if itself. People often behave this way for a reason. You're liable to for the same reason.

I read the article and thought to myself that there's not a single thing in there that I don't recognize in my coworkers. :) Fwiw, I think articles like there are really bullshit. They just make self-doubting people doubt themselves even more and the real assholes never understands that the advice applies to them.

There's a common quote about an expert being someone who has made every possible mistake in the domain at least once. So you may be on the right path!

As a solo developer who does all these, I have a lot of self-loathing.

The worst are the "What the hell was I even doing here?!?!? / What does this even mean?" moments.

Like at one point I thought this code was obvious and worked great. I remember feeling that way about it and feeling all accomplished here, but now I don't feel that way and this code seems impenetrable or at least a big hassle now...

Software development is the continual discovery that you were an idiot two weeks ago.

God knows.

Especially when you're new like me. I go back even after a month and "Oh man I'd never do it like that again."

No to say it didn't work or was even bad but I just found a better pattern to handle that case that I'd use by default now and it changes so much.

I cringe when I look at some code I wrote. Some of it is maintenance nightmare, however, I read a book and found a better way to do some of those things. With each day I get better at writing maintainable code, I do less of duct taping to make things work.

I feel like I'm actually getting good at this software thing, and I'm actually not a total idiot. However, sometimes I look up simple things and I think I'm still an idiot. But when I look at the bigger picture I can see how much I have learned and understand the whole process. Servers, database, software, documentation, good code is great but there is more to software and I love it.

My advice: see if you can rewire your internal rewards. That feeling is the acknowledgement that you learned something.

The fact our ancient rewards system considers it embarrassing is of little importance.

Thank you, I'll keep this in mind for the next time it happens.

It's just that the code is still there, and will need to be replaced in due time. Over and over again.

There's no feeling quite like going 'what the hell was the person who wrote this thinking??' and then doing a git blame and making a humbling discovery.

Software developers often forget that their biggest critic is their future self.

#9 Flippant git commit comments nobody can understand a month later.

Even future me hates me.

I’ll add another one: Nitpicking constantly.

I’ve worked with people who will fight and argue over every. Single. Detail. It drives me insane and usually makes me quit faster than anything else when I encounter those people.


There are 2 types of developers: those who enjoy philosophizing about programming minutiae and then there are those who enjoy shipping products.

Don’t get me wrong. There are times when details should be discussed, but not when it’s coming out of some desire to engage in a power struggle and show everyone else how smart and in control you are.

There's a subtype who ignores your review request for weeks but delivers a verbal drive-by nitpicking, as he's leaving work early, which focuses on trivial matters of opinion and ignores obvious bugs in functionality.

Not just nitpicking either. It's enlarging a minimal problem to make it a major problem that must at all costs be solved.

How everyone can be right, but everyone gets annoyed anyways.


Developer A, a younger developer who is recognized company-wide as an expert in the latest features of blb++.

Developer B, an older developer who maintains widely used legacy code in blb++ involved in about 80% of a billion dollar multinational's profits.

Dev A: Your group should change the coding style in this application to the new standard.

Dev B: Sure. But since we don't have good Unit Test coverage in this area, we shouldn't just go in and do it by hand. I need to guarantee correctness. I've done a change this big before using a syntax driven automated code rewrite tool, and one can guarantee correctness if the code structure allows it. If you know of such a tool for blb++, you can help me out and we can do it.

(Background for the reader: No such tool exists for blb++ which can guarantee correctness. Such tools do exist for other languages in Dev B's work experience.)

Later on, Dev A is grousing about how Dev B is just coming up with esoteric excuses to do nothing. She's kind of put off. Dev B is put off, because he's kind of been accused of making stuff up, even though he's talking from a particularly interesting and valuable part of his work experience.

Dev A is correct, in that the new coding style is better, more efficient, and would even prevent mistakes going forwards. Dev B is correct, in that he really needs to guarantee correctness, especially if he's going to make a change that big which has no visible benefit to stakeholders. What's the takeaway?

> What's the takeaway?

"Unfortunately this tool doesn't exist for blb++, but we should still improve our coding style for this application. Is there another way to improve our correctness testing? If we allocate some time to improving unit testing, maybe we can do this refactor once we have X% coverage in these critical parts of the codebase."

I.e., acknowledge that both goals are valid, and look for alternate solutions to make incremental progress towards both. Finding middle ground is unlikely to be satisfying from either viewpoint. In this example, Dev A isn't getting a quick win from fixing the coding style right away, and Dev B needs to allocate some time to improving testing. But especially when working on big teams and projects, this kind of piece-wise progress is where I've seen most things get done.

This kind of solution might be pointed out by either Dev A or B, or they might need to bring in a fresh pair of eyes if they're both already frustrated. :)

Devs A and B collude to convince manager M to fund development of the nonexistent tool. They use it to update the code base with confidence. Dev A supports the new code. Dev B supports the tool. Dev B and sales S team up to sell licenses and migration support to other companies that use blb++. M pockets a nice bonus, and gives A and B 2.5% raises. When the corporate well runs dry, S moves on to something else, and B open-sources the tool. A and B get invited to blb++ conferences to earn peer recognition and prestige. They co-author a book, and then pocket a bigger windfall than M's bonus. A and B get hired by another company to support a fork of their tool. Then blb++ loses popularity, B goes into consulting, and A becomes the B' to someone else's A'.

Dev A wins because management is trying to fire Dev B. Dev A works longer hours and is paid less because they're young and stupid and don't have a family. Dev A is better for business.

Dev B gets laid off, and Dev A works until burnt out.

Game over. Everybody lost except management.

Dev A is a junior developer that doesn't understand the time and opportunity cost of rewrites. Or that projects and managers don't care about style, they want new features and stability. Doing things better going forward is great. Going back to change legacy code for style? Frankly, that's a big waste of time, and management will not look fondly on your lack of visible or impactful progress.

I was Dev A a couple of years ago. Don't be Dev A for too long.

It can be done in a way that show massive visible progress and impact. Just have to fill plenty of tickets for doing plenty of changes. Adding new features that are already present, sometimes renaming the file that had them if necessary. Fixing bugs, possibly freshly reintroduced because the existing fix was purposefully ignored or removed.

Do you know that progress report and commit messages don't even have to match what was actually done or committed? The project is your playground.

Uh, Dev A should take the time to understand Dev B's argument before writing it off as esoteric excuses.

>>> the new coding style is better, more efficient, and would even prevent mistakes going forwards.

Except that's wrong. The new style is not "better" at all. Replacing CamelCase with underscore_variables and forcing brackets on new lines make no practical differences.

All the time to rewrite hundreds of files was just churn.

Except that's wrong. The new style is not "better" at all.

RAII is tons better than older memory management styles in C++.

The newer styles of C++ initialization are tons better than the older ones, which seem to inevitably result in a dozen ad-hoc mini-initialization frameworks.

Replacing CamelCase with underscore_variables

That is indeed arbitrary and useless.

forcing brackets on new lines

Forcing brackets on new lines is useless. Mandating the use of brackets has been shown to be very useful for reducing bugs.

In my experience people provide excuses because they're being addressed by management in a way that is unnecessarily negative and/or management is wanting to find "guilty people" and not work out which is the best person for the job of fixing things.

That the article is clearly refusing to even mention this casts a big shadow over the rest of its conclusions. Some of the other things are the result of personality (like taking credit or sloppiness) but making excuses is 100% context and a defensive coping strategy literally every time I have seen or done it.

Yup, "no excuses" is a stupid attitude. Say I wanted you alone to get me to Mars tomorrow. It's absurd to say "no excuses" when you fail. It's possible that you're making understandable demands, or they just don't know how to do it faster.

>Most developers are enthusiastic people, but sometimes you may have the chance (or misfortune) to work with a negative one. Negativity is infectious

Western society has a weird optimism bias. I think it goes hand in hand with extroversion bias, and it keeps many people from being genuine.

>those developers are clearly at the top of Mount Stupid

Frankly, that's insulting. Being a negative or critical person does not make one stupid, on the contrary, I would argue that the skepticism that comes with negativity is an asset for engineering. This advice is carte blanche to dismiss criticism.

Sometimes we need to remember that positive interactions make people happy, and negative ones make people sad. And that's what they'll remember about the interaction afterwards.

For a lot of people, these basics apply even to technical discussions. Criticizing their code makes them feel bad, and so you are a jerk. Meanwhile, you think you are criticizing their design to help them improve it. This is a mistake on the part of the person receiving criticism, but it's very rare to have a manager or mentor who can point this out to them, and help them dissociate their implementations from their self worth.

Couple into this other culture's aversions to even just saying "no", and we have a complex and challenging world to navigate. I'm constantly in search of the team where we don't have to do silly social dances to make sure everyone feels valued, and can just criticize and fix, since the members know they're valued by virtue of being part of the team to begin with.

But that is asking a lot of people - they have to have reflected on these issues. Since this is rare, and the larger a team gets, the more likely you'll have to deal with those who haven't dissociated yet. Makes for a great mentorship opportunity if you can find a way to have them "let you in".

True enough, but it's not just about making anyone happy.

Even thick-skinned people have a hard time with another type of negativity:

"Negativity" doesn't only mean high quality criticism, or pessimism.

It also refers to low quality, demanding, draining, and often competitive and repetitive discussions where critics are full of it.

Sometimes there is also bullying, always talking over, making out that coworkers are stupid, taking credit for others work, not pulling their own weight, trying to trick people, manipulating people to their disadvantage, insinuations, word-twisting and game-playing, that sort of thing.

That sort of negativity is draining. It's negative in an energy and morale sense.

That's not about hurting anyone's feelings, and it's not even about sadness and happiness.

It's just that some ongoing interactions drain everyone's energy without giving much back in exchange.

Everyone does it occasionally. Some people do it a lot - those can destroy a team and a company.

> "Being a negative or critical person does not make one stupid"

It does if you have engineering goals and the _manner_ in which you are being a negative or critical to other people is preventing the team from achieving those goals.

Negativity does not always come with skepticism, by the way:

Some people are negative in the sense that they demoralise others, yet are full of uncritical enthusiasm to use every random idea and technology they hear about. Which may be what is demoralising others :)

I thought I would make an addendum:

> We all know at least one developer who ... will always insist on following “best practices” without understanding why those practices are considered “best” (there is no such thing as best practices that adapt to every team)

For any developers out there recently starting on your path, I cannot stress the importance of taking the time to seek out best practices for new skills and technologies you pick up.

Find all the "best practices" you can, then ignore some of them for a while with some pilot code, then try to follow a combination of the ones that make most sense to you. Compare the two modules for legibility, clarity, maintainability and conciseness.

If, however, your team already has an entrenched way of doing things, don't be that jerk who comes to the party blasting their own music. Follow the existing style guides to a T and only offer advice for improvement after you've spent some time working with them.

Even if you immediately recognize a glaring problem in the way things are done, no one is going to take you seriously and you'll just come across as arrogant unless you already have rapport as a perceptive and helpful team member.

Take time to understand the culture around you, but don't, as this author suggests, ignore the tried and true devices of other cultures.

I agree.

I'd like to add, in my experience some developers firmly (and sometimes loudly, judgementally) join a working environment and espouse best practices they've acquired from a cultural background they've worked in so far. Or read about, or watched a training course about. And they don't know (or care if) they are pushing things which are advertised as "best practice" in some domain or other, but aren't half as universally agreed upon as they think, and aren't half as effective or appropriate as they think in the new job.

My point is: Some "best practices" aren't as universally agreed upon as people think, and people are often thinking in a bubble.

If you're at a new job, and you see a glaring lack of what you've learned is "essential" best practice, I'd urge caution in assuming your new team are ignorant or that your managers are as clueless as you think at first, even if it looks messy and disorganised.

Of course they might be clueless! But it takes time and deep questioning to be sure.

Of course your experience should be brought into each new place you work. You're hired to bring in what you know, not just to fill a seat. By all means talk about your experience, about things you have actually done which worked well, and about what industry leaders are currently talking about.

But if you feel the need to "teach" everyone straight away how to work better, give it time and consider the possibility that people might have given it considerable thought and experience of their own. They might even be familiar with much of what you're talking about, and rejected it or found a different approach. (Or they might not - that's to be discovered.)

I say this because I've seen people turn up and, in effect, try to start fights long before they have spent the time to figure out (a) others in the team are quite experienced and familiar with the same practices but have decided on something else, (b) different industry bubbles actually do have different best practices for similar problems, and (c) they can't see some kinds of development strategy that are in use, because subtlety.

It's very weird to describe arrogance as a problem only for people near the bottom in terms of ability. Arrogance is an issue at all skill levels.

This, so much. It's where review comments like "why don't you just do X?" come from, where "X" is a systematic change. Condescension coming from a place of strength doesn't make it okay.

I was just imagining the opposites of these and remembering some of my favorite people I ever worked with. I've only ever worked with 2 developers who I didn't like, and they fit this list almost perfectly.

Be humble, be kind.

Individuals often have little control over these things, and they are indicative of deeper cultural incentives. Without transgressing some, nothing will get done in many orgs. I see it a lot in big money VC backed startups, much less so in bootstrapped businesses. It’s important to be non-judgmental of people doing these things and simply leave or master Machiavellian moves to clear a path to promotion as other players are, eg. via optics or rallying support to oust “toxic” rivals for the good of the team.

I agree. The system in place is to blame, and those people rose because they are just playing the game that is dealt to them.

I'm not so good at the Machievellian political aspect, so I personally opt to leave when I notice behaviors like this are rewarded in the current system. I also wonder if I was able to rise in an organization that functions like this, if I would b the type of person to actually root out that toxicity in the first place.

You won't be able to because the people higher are doing the same thing. Unless of course you make it to the top.

Game theory suggests this is true. Dog eat dog. But do you think this is because incentives are misaligned, or do you think that even successful bootstrapped businesses, if around long enough, will fall victim to this culture?

>> will fall victim to this culture

The issue with this culture is that using optics to achieve an advantage generally works, but is not a good long-term--big-picture strategy.

So reasonably, of course you need to learn all of these quirks, just to be able to get rid of the vermin in your org.

I agree if we decouple optics from genuine promotion in which an employee "manages upward" their true successes. I know that people are more nuanced than this and it's not black and white. So this will become quite tautological.

As mentioned above, either the code works or it doesn’t … but it needs to work in combination with all the code being added to the codebase by your teammates.

That is Linus' job (or one of his trusted lieutenants). Kernel contributors generally do not interact with each other directly. There are simply too many kernel contributors to hold all kinds of "team meetings". The problems that the article mentions, exist mostly in a corporate environment. Virtual, distributed teams generally do not have these problems.

Software engineering is probably the most collaborative work in today’s world.

Yes, but successful software, such as the linux kernel, is not developed in the way he believes it is. What corporate IT typically does, is certainly not a reference.

A manager who doesn’t understand this is a manager who doesn’t understand software engineering.

A manager who does not read, evaluate, and merge pull requests, is not a manager; and does indeed not understand software engineering. The article complains about the side effects of one particular, rather outdated approach to software engineering. You will, for example, not hear the PostgreSQL or the Debian team complain about that kind of issues.

What corporate IT typically does, is certainly not a reference.

The article was in no way specific to "corporate IT".

But, literally by definition -- we're talking about the huge mass of applications that run most of the planet, here -- overall, "corporate IT" is way more representative than the handful of elite open source projects you're referring to.

From my experience at various companies, "technical managers" fit well into most of these categories. Because actively coding and managing a team of multiple developers can bring most of these issues out of you. Example: "getting away with your architectural decisions because you are presenting your ideas with a managerial authority and ignoring better solutions presented by other senior developers."

Good article. However, I'd like to talk about documentation and comments, as they were mentioned a lot.

I think the industry should rely on them less. The trouble is, nobody updates every comment, and documentation is even more out of date. They cannot be automatically verified either.

Getting worked up about it is just going to give you stress. Further, it is terribly subjective. Instead, advocate for the other good practices - variable names, clear separation of concerns, single responsibility, functional purity, etc.

I have this inkling, actually, that the more comments there are in a function, the worse it is. Even high level overview comments, like some outlining an algorithm, could instead be high-level code, composing the various pieces together.

It is very rare that a concept is actually nuanced in code. Instead, it is usually the implementation - and you can improve it.

That said, it's a ton of work, and we have to ship a product at the end of the day.

So here we are again, discussing tradeoffs in tech, and I think in these cases a comment block is an OK compromise. Just be cognizant that it IS a compromise.

Common misconception, but that graph does not explain the Dunning-Kruger effect. It's actually more like a linear curve, it's just a little flatter than the graph of y = x.

The Dunning-Kruger effect is that the best people underestimate their relative abilities, and the worst overestimate. So, the worst 10% may estimate they are about average, while the top 10% might estimate they are in the top 20%. These are more realistic figures than the graph used in this article.

People try to shoehorn Dunning-Kruger into situations where it doesn't make much sense.

Dunning-Kruger boils down to something a lot like "people tend to think other people are like themselves". So if you are low-skill at some task, you tend to think other people are low-skill, and as a result you see yourself as roughly average. If you are high-skill at some task you tend to think that other people are also high skill, so you over-estimate the average skill level.

This article is an example of shoehorning it in when it makes little sense, and I argue that they are claiming that Dunning-Kruger says the opposite of what it actually does. Yes, people can be arrogant and incompetent, but they can also be arrogant and highly competent. Dunning-Kruger would imply that someone who is highly competent would not be likely to be arrogant, since they would overestimate other people relative to themselves, and as a result they would give more credence to the ideas of others than is warranted given their relative skill levels. A total incompetent would indeed overestimate their own ability, but they would still see themselves as roughly average, which you wouldn't think is likely to be the trigger for an arrogant attitude.

All people will from time to time feel like the ideas of another person are silly or not worth listening to, but of course it is considered extremely rude to express those feelings. I think arrogance is just the expression of those feelings, which makes it a symptom of poor socialization.

> So if you are low-skill at some task, you tend to think other people are low-skill

Dunning-Kruger effect is not a way of thinking or reasoning about yourself or others, but a cognitive bias. You need to be put into a quick decision making mode where you have to compare your skill to others for it to skew your assessment.

Love the article except for the part that encourages developers to rat on other developers. No. Unless another developer is so bad they are putting your job in jeopardy, let them be. It's the manager's job to sort that out and do something about it.

Throwing other developers under the bus when ones own job security is not threatened should be the number one thing on this list of asshole things a developer can do by far. By very, very, very far.

This writer stole "Mount Stupid" from Zach Weinersmith (Saturday Morning Breakfast Cereal) without crediting him! http://www.smbc-comics.com/?id=2475

Example of the arrogance mentioned in the first point: I have met folks who have seriously argued to me that their code contained no technical debt because "We wrote it right the first time." The code at the time was something in the range of 50,000 lines (so assuming 30 lines of code per page, 500 pages per textbook, you'd have to read ~3 textbooks on the subject of what their code does to understand it) and contained no test suite... and maybe 5% of the code was comments.

Sloppiness seems to come with enough examples... but I must say that it's not really a direct problem unless you are in a supervisory role and nevertheless insist on modifying the codebase, as that makes it significantly harder to correct you on your sloppiness.

Disrespect of others' time is being directly connected to meetings here -- I sympathize but I do think that the bigger issue is that "deep work" is best done in ~2-hour uninterrupted batches, and so a company culture which encourages actively scheduling those on shared calendars so that we can have "open spaces" for meetings during other parts of the day, would help a lot. Especially, I am growing more confident that meetings which exist should revolve around some decision to be made that needs input from a bunch of people -- in other words, every meeting is a negotiation and if it's just a "progress update" or a "question and answer" session it should be moved to an asynchronous medium like Slack or (to a lesser extent) email. If you insist on daily standups at least have the courtesy to schedule them in the afternoon so that when I come in off of my morning commute having thought, "I am going to do X, then Y, then Z" I am not burdened by "I have only 1 hour to work on X before I have to drop everything for the daily standup..."

I am probably more guilty of the constant negativity, I think a piece of wisdom from Seth Freeman is helpful here: that one wants to separate problems from people and be hard on problems, soft on people. You can be constantly be negative towards a problem and this will be tolerable if you are consistently cheerful towards the people who might have other needs with which they pursue those ends. "I am really worried that without a proper auth strategy we may get hacked, I know that you all strongly value being able to go forward without wasting time on such a frustratingly difficult problem, I fully understand that, but there has got to be some way that we can get a proper auth strategy which doesn't bog us down so that we're also not trivially hackable" is a very negative position but it somehow doesn't carry the same "drag" as "you're so stupid, trying to implement this insecure thing, you're going to be the reason we get hacked."

Greediness is a hard thing, definitely, but I would observe that all of the examples seem to have to do with private communication channels, and I wonder whether that's endemic to the situation. I also wonder how it subdivides with a manager taking credit for the successes of their team -- in some cases this respect is due and in some cases you feel like "we spent more time evading my boss than being led by them!"...

I think the weakest part of the article was "disregard for the team," I feel like that's just a catchall for "doing anything that was annoying to me personally" and it's like "well yeah but that's not helpful." I think any friction can be couched as "disregarding the team" whereas true disregard has something to do with "you went off and made your own decisions and never told any of the rest of us about it and we could have told you that they were not wise decisions because of needs that you would not have been expected to anticipate" -- but the sin there is really just falling out of step with the community and thinking "I can sail this ship entirely on my own!" and I think it's less disregard-for than lack-of-community-with. Like it's great to have ownership of some problem, but keep in constant conversation as well.

Lack-of-focus is I guess an irritant but I've never worked with someone where it was so bad as to "hate to work with you"... I think some of that is a lack of leadership-focus, if you have an aggressive timeline to deliver an aggressively minimal product then there is very little room to dither? But that may just be that I have not worked on a hard enough problem where one needs to upfront a serious enough amount of investment to force such dithering.

I was also frustrated to see "lack of accountability" defined as "more focused on making excuses than solving problems" because to my mind those are two separate issues, the "I am never to blame because I will point the finger at someone else!" is toxic, but it doesn't become less toxic if someone is like "yeah I mean I was just doing my best with the API results that Phil was giving me, Phil really screwed us over with this one, let's solve the problem by creating an API v2 that doesn't have his shitty endpoint in it, instead the endpoint will work like this, and then I can write code that's actually correct." That revolves around a solution while still having the attendant point-the-finger-and-blame attitude and I don't think that the solution-focus really removes the awfulness of the negativity.

Kind of my take aways are:

1. Keep learning and growing, shun practices which set you up as someone who knows everything and has nowhere to grow.

2. Keep honest with yourself about what's really going on, you can massage the description to others but you should be clear on 'we are not meeting this deadline because our contractor is two weeks late delivering this component and we cannot start building this next part without it' -- use that clarity to try to creatively evade those restrictions in the future, 'can we agree on a black-box specification so that you are not blocking my peer developers from working? Like, here's an example JSON file or two that your API will return in response to this query?" -- Do not lie to yourself in cases of "I am screwing this up" or "I am overburdened" or whatever, as those lies breed bigger lies later, that developer who is like "It will be done this week!" for weeks on end.

3. Keep compassion in your heart for everyone else. I am trying to separate that empathy as much as possible from these other criteria and just dump it in one. Do not be aggressive or negative or so with others, they are not the problem, they are not to be recipients of blame... they are fellow human beings who need to also grow and be respected just like you do. Don't say anything which you would not want someone to hear if you said it to their face, for example.

4. This was not really discussed explicitly in this article, but don't settle for mediocrity. Go out on a limb, take a risk, try to do things that might fail. I think one of the things that can really make me hate to work with you is that you would rather copy-paste some convoluted solution that worked once to a problem with considerable limitations, rather than that you're really trying new things, refactoring, simplifying. One thing I perversely love to see in source code is a brute-force approach. Just an "I don't know what the right way is to solve this but there are only 8! different ways this can happen so I'm just going to iterate through the 40,320 of them and see which one is best, we can improve this with some heuristics later." But it has to be an interesting problem of "which of these things is going to be most effective for our users?" for me to have that respect. You stuck your neck out and did something that other people would have just shrugged and said "that sounds impossible, let's move on" and I love that spirit.

5. And finally there is some sort of pride/humility thing going on, don't think that you are the center of the universe and that every other developer exists to make you effective, but rather give up your ego in service of the art that you are creating. If you understand this code as a vehicle for yourself to be validated then I will probably not like working with you, if you understand this code as a joint effort of love and service, I will probably call you my brother or sister.

People love to work with me, but managers love all of those flaws.

Did you mean "People hate"?

No, I mean I don't get promoted for trying to hear what other people say, while managers can't say enough nice things about the hipster-ass who breaks everything he touches trying to turn it into GraphQL no matter what.

But I didn't write that very clearly the first time.

Re: documenting code, I've gotten a lot of conflicting advice over the years, with some people insisting code should be documenting, and others insisting on docstrings.

Best to have a balance, I think:

For loosely typed dynamic languages, docblocks on interfaces (function/method definitions, interfaces, etc) are a good idea.

For languages that require types in their interfaces, it is helpful for larger units(classes, functions, methods) to have a summary comment.

Inline comments are still very useful for explaining how and why of implementation details.

While I reject the stance that you should never have comments, I also see cases where better names and simpler logic can alleviate the need for more comments.

This sounds more like how to make anyone hate to work with you.

I'm in healthcare and this translates wonderfully into my field. Especially with being one staff short than have all, and one be negative.


Edit: The domain is unreachable for me, I assumed it was for others also. It is still unreachable for me for some reason.

Ironically this domain is blocked at work for being a "spam url".

'Arrogance' is more often than not mostly a matter of perception. Truly arrogant people can seem nice - so we generally don't want to ascribe negative qualities like 'arrogant' to them.

'Real arrogance', and even greed, backstabbing - those can also done by people who seem to be well liked.

'Perceptive arrogance' I think is mostly a matter of posture, demeanour and communication. If you smile, let other people talk, have an easygoing manner, and are agreeable, you will not be perceived as 'arrogant' even though you may have all of the qualities of a truly arrogant person.

I don't think 'perceived arrogance' has anything to do with actually humility or gauge of one's own abilities, or of one's sense of self importance in the group.

If you're terse, blunt, dour or gregarious ... it can be perceived as arrogant.

'Real arrogance' i.e. the notion that one's thoughts and ideals matter more than others etc. I think is not even correlated with posture and communication style.

The most successful people in the corporate world are pure political players, and have never cared about outcomes, doing a good job - anything. All they care about is perception and their careers. But they are actually nice enough, generally charismatic.

Actually caring caring about a product can simply cause contention, and possibly give the perception of arrogance.

I prefer to consider arrogance in terms of measured behaviours, outcomes etc.. Glib political climbers to me are arrogant. Anyone actually trying to 'do a good job' and stepping on toes is just a bad communicator.

A lot of good nuggets in here. Dunning-Kruger and Mt. Stupid is one, but I especially like this quote "meetings are just scheduled interruptions".

Referencing Dunning-Kruger effect in any topic on software engineering is very arrogant and pretentious on itself. This says more about the author, than other people's stupidity. The article is just bad.

Also the "Negativity is infectious." feel-good bullshit.

"If someone complains, it focuses the attention on the negative side of things."

(...after a whole article complaining about behavior s/he does not like, ironically)

If you want to see big disasters look in organizations where warnings are ignored as "negativity" and people blame the messengers.

Also, sometimes relentless negativity has some fringe benefits (moderate perfectionism). There's pros and cons. Just don't share that attitude with your coworkers. Hold yourself to a high standard, but be more relaxed with others.

why does it "say more about the author"?

does your comment about Dunning-Kruger "say more about you" than the author?

Well, if you use a fictional interpretation of something known as a justification to claim that other people are stupid - you provide nothing of value regarding other people. And the only thing that can be extracted from this is that you feel the need to claim that other people are stupid.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact