I get angry when my coworkers make mistakes

by Alison Green on October 2, 2024

I’m off for a few days, so here’s an older post from the archives. This was originally published in 2017.

A reader writes:

I’ve got a question regarding how much “mistake tolerance” is expected in the workplace.

Just to give you some background, I’m a (tech) team lead, which, in my case, means my daily job is not very different from that of other team members, except for the part that I get to make technical decisions concerning the projects we are doing. That includes deadlines, technologies, methodologies, features to be included, etc. and most importantly, I decide whether a piece of work by any team member is acceptable. However, I don’t “manage” people; that is, I don’t give time off, I don’t give them feedback, I don’t decide their raise, etc. There’s a manager to do that.

Now to the main question. I have very low, almost zero, tolerance for mistakes. Whenever I see a mistake in anyone’s work, especially trivial ones, I will get very angry. The rationale in my head is always “We have ONE job and one job only, and that’s to get this done! No excuses.” As such, I will remove the person from the project, in addition to having a detailed (sometimes heated) conversation with both the person and our manager on why such mistakes are not allowed in my team.

So how bad is this? I know my intolerance could probably be attributed to some sort of OCD, and sort of know it is not good. But I just cannot forgive mistakes easily. Do you have any advice?

Yeah, what you’re doing sounds pretty bad.

I see two issues here: First, your expectations about normal amounts of errors are off. And second, you’re taking it really personally when mistakes happen and you’re having an emotional reaction where one isn’t warranted, rather than handling it professionally. (Which, as people are pointing out in the comment section, is a mistake in itself! So there’s some irony there.)

On the first issue, people are going to make mistakes because you work with humans, not robots, and humans make mistakes. If someone makes a mistake occasionally, that is normal — and you should see it as normal and not an outrage. Perhaps you’re the very rare person who truly never makes mistakes in your work. If so, you’re something of a unicorn. That’s not typical. If you are that unicorn, good for you — that’s a rare talent. But if you want to work with other people, you have to recognize that you’re not normal; if you expect others to be unicorns too, no one will want to work with you, because you’ll be out of touch with reality.

Now, obviously there’s a point where someone is making too many mistakes. And that brings us to the second issue, which is how to handle it when that happens.

Right now, you’re reacting very emotionally: you’re getting angry and having heated conversations. There should rarely be any need for that at work, and by doing it, you’re almost certainly alienating people and making no one want to work with you. That’s a big deal — not only are you making working with you a bad experience for other people, but you’re also impacting your own professional reputation. That will matter when you’re looking for a promotion, a raise, or a new job, or even just when you want to be included on something that other people don’t want to work with you on.

Here’s the thing that you’re losing sight of: At work, you have the tools you need to solve problems calmly and rationally. Getting angry and emotional says to other people that you don’t know how to do that. It makes you look out of control, and it can make you look inept. You don’t want that.

Your goal needs to be to solve the problem, not to punish people or let them know how wrong they are or how much they frustrated you. Instead of having a heated reaction, you just need to deliver information calmly and clearly.

That means that if someone makes a single mistake, all you need to do is say something like this: “I found mistake X. Can you take a look at it and fix it for me today?” If relevant, you can add, “Let me know if you’re not clear on what I’m talking about and I can walk you through it” and/or “Can you figure out how that happened so we can make sure to avoid it in future rounds?”

And if someone makes mistakes regularly, that’s a pattern you need to talk to their manager about, since their manager is responsible for addressing it. And that should be a calm, matter-of-fact conversation — as in “Fergus is regularly making mistakes like X and Y. I’ve pointed it out to him, but it’s continuing to happen and I’m concerned about the pattern. It’s causing me to have to redo his work and making me reluctant to keep him on the project.”

But there’s almost no reason to ever have a heated conversation over a mistake. This stuff shouldn’t be so emotional.

If you find that you can’t control your emotions about mistakes, it’s probably worth exploring with a competent therapist — because a pattern of strong negative reactions to something that doesn’t warrant that intensity is usually connected to something more deeply rooted, and likely isn’t about work at all.

You may also like:
{ 226 comments }

{ 226 comments… read them below }

  1. Foreign Octopus*

    Reminds me of the guy who tanked an interview because he never makes any mistakes. I feel those two would have fun together /s

    1. ReallyBadPerson*

      How do we know this isn’t the same guy, now (finally) employed after convincing an employer of his perfection? I mean, have this LW and that one ever been seen together IRL?

      1. Hlao-roo*

        Ha! This letter was originally published August 2017 and the “I told my interviewer I never make mistakes” letter is from February 2024, so it could be that he got so fed up working with mistake-prone coworkers (a.k.a humans) that he decided to try to find a new job.

        Can’t say I’ve ever seen the two of them in the same room at the same time :)

    2. Emotional support capybara (he/him)*

      Came here to say this, lol. Unstoppable force, meet immovable object.

    3. HelloBonjour*

      It looks like more than a coincidence that AAM reposted the particular letters on consecutive days!

      1. H*

        In my most recent job, I was on a team of individuals who were all like this. I was the mistake maker and was terminated after about six months. The termination was likely fair but the rude comments and condescension only exacerbated my mistakes and made it much harder to ask for help and support when needed. What my team didn’t know is that I was in the diagnostic process for a rare progressive nueromuscular disease similar to young-onset Parkinson’s or MS. It was fair people were frustrated but being rude and aggressive helped no one and only made me feel far more scared of the future.

        3. KJC*

          I’m sorry to hear that happened to you. And your comment makes a really good general point – being overly harsh about mistakes, as opposed to calmly focusing on problem solving together, makes people afraid to bring errors to your attention when they notice them or to ask for help when they need it.

          1. Jackalope*

            In addition to that, it can put you into survival mode if the person pushes your button right, and you’re flooded with fight or flight hormones, anxiety, or whatever, depending on your reaction to said harshness. Rattling someone and pushing them into that stage causes MORE mistakes, because you can focus on something properly if your body is in panic mode.

          2. Working Class Lady*

            It’s pretty bad, and likely this LW has driven away a lot of people and actually hindered progress. Imagine a workplace where you could literally never bring up a mistake to your manager – ever – without them losing their temper.
            Small, otherwise easily fixable mistakes will go uncorrected.

            At least this person is somewhat sepf-aware but they need to control themselves ASAP or honestly they aren’t cut out to manage people.

            2. Great Frogs of Literature*

              The worst coworker-caused production issues I have seen were not the original mistake, but the instinctive panic to cover it up. People would have been a little annoyed by the original mistake, but it would have been accepted and normal, and no blame would have been assigned. As it was, there were incident reports and several serious conversations to the effect of, “When you realize you’re in a hole, you need to STOP DIGGING and ask for help.”

              Nice guy, but he was kind of exhausting.

    5. I Would Rather Be Eating Dumplings*

      Oh man, this guy seems to have at least some degree of self-awareness; he acknowledges with the end that it’s a problem (‘so how bad is this?’) and that his intolerance of mistakes might be rooted in something other than reasonable expections.

      That interview guy…IFRC he felt throughout that the issues were entirely not with him.

    6. Curious*

      Reminds me of TOS “The Changeling”:

      I am Nomad I am perfect That which is imperfect must be sterilized.”

    7. Working Class Lady*

      LOL.
      It’s ironic because people who believe they’re perfect have no incentive to double-check their work, to adapt, or to improve themselves or learn.

      I’ve seen it all the time; guarantee the “Never Makes Mistakes” LW makes them all the time and either refuses to admit it, OR blames others.

  2. Hlao-roo*

    This person should hire the letter-writer from “I was rejected because I told my interviewer I never make mistakes” (posted February 13, 2024)! I think they could work very happily together.

    1. dulcinea47*

      Try seeing people as human… people don’t have “just one job”. They have a million things to do and only one of them is their job that they’re paid for. Expecting anyone to be laser focused and perfect(!) all the time is a terrible standard that no one can meet.

      1. Overthinking It*

        Yes, and think about the part where LW gets especially wound up about TRIVIAL mistakes. Workers may be concentrating on the more crucial issues, and let little thinks like typos or page numbering get past them. I the lead wasn’t so busy dressing them down, maybe they could work our a system for catching these little errors, as well as the big ones.

        1. JSPA*

          I figured it was coding, in which case typos are not entirely separable from substantive errors, and running your own code to check it / general debugging actually is a core aspect of the job, and triple, quadruple and quintuple checking, line-by-line and character-by-character is not an unusual expectation. (The emotional intensity is still weird, and I can imagine people making minor errors just to get off of any team that answers to this LW, if they are this…intense…about everything.) If it isn’t coding, the response is more egregious.

          1. ThatOtherClare*

            You know, you might be on to something there. Perhaps the letter writer feels like they’re drowning in incompetent people because everyone is taking their first opportunity to make a non-firing-level mistake in order to switch teams. The old ‘whoops I accidentally deleted a semicolon, oh dear‘.

    2. Red Wheel Barrow*

      To be fair to this OP, they recognize that their response is not great and may stem from their own mental health issues. They’re also asking for advice, unlike the other poster, who was asking for justification (and doubled down in the follow-up, if I remember correctly). It’s true this OP’s behavior is very bad, probably worse than they realize, but it sounds like they’re potentially open to hearing what Alison says. I don’t want to pile on when they’re asking for advice that could improve their behavior.

      1. MigraineMonth*

        This is a good point. Just because these are reprints doesn’t mean the LW isn’t reading the responses, people. Let’s all remember to stay constructive while Alison’s away.

        1. Ann Onymous*

          Sometimes being constructive means being honest!

          True, this isn’t AITA-Reddit, but some of the responses there to people who are arrogant/cruel/selfish, etc. would peel the paint off the walls – the commentariat lets those people have it in no uncertain terms. And a surprising number of LWs later write follow-up letters saying that the blistering comments were RIGHT – that they made the LW sit up and take notice, look unsparingly at their own behavior and make positive changes in their treatment of others. All because the commenters pulled no punches!

          Sometimes, soft-pedaling is worse than bluntness; it can lead a person to coast along doing what they’ve always done, getting what they’ve always gotten and wondering why it never gets better. Nobody wants to see people cursing out the LWs in AAM, but please bear in mind that those LWs are writing in because things are NOT going well for them and they WANT advice and help! Advice and help, folks – not coddling.

    4. Sunflower*

      I thought it *was* that letter but knew something wasn’t quite right. I guess I got mixed up because they have the same vibe, except on different side of the desks.

    5. Ms. Eleanous*

      Did it never occur to OP to build in an editing layer? Tom edits Rachel’s work, Rachel edits Linda’s, etc.

      Gee, OP, i would count that lack of an editing system as your BIG mistake.

      1. LinuxSystemsGuy*

        This is software development, that’s literally enforceable. Most places I’ve worked the dev team submits code to a branch repository, but it has to be reviewed before the branch can be merged. Sometimes it’s a specific “Bob reviews Sue, Sue reviews Eric, etc…” other times just “someone must review your code”, but in any case there’s an enforceable process.

        2. I am Emily's failing memory*

          It sounds like he might even be the person doing that second-layer review (“I decide whether a piece of work by any team member is acceptable”), but his expectation is that he should never actually catch anything needing revision in any code he reviews, because it should be error-free before it comes to him.

  3. T.N.H*

    We now have proof that even robots make mistakes because ChatGPT continues to make stuff up out of whole cloth. I doubt this person is that unicorn. They made a huge mistake in handling this situation, which makes me think they just don’t recognize their mistakes (or blame others for them).

    1. UnCivilServant*

      I have to be pedantic, ChatGPT isn’t a robot either. It’s a large language model. Just a bunch of software mashing together tokens it doesn’t understand and numerically rating how well the correlate to its library of token associations.

      1. T.N.H*

        I mean, you don’t. It’s already being used in robots who then presumably make the same mistakes. But more importantly, I read Alison as using robot colloquially.

        1. Dawn*

          I’m trying to be nice here but at the same time, yeah….

          Every year of my life I get increasingly frustrated by the existence of people who insist on unnecessary pedantry. It sure doesn’t make you any friends.

          1. Alpacas Are Not Dairy Animals*

            With something like ChatGPT I think being very precise about what it is and isn’t is worthwhile, because a lot of the hype is based on overblown promises about what it is and can actually achieve.

      2. MigraineMonth*

        I’m sure someone somewhere has created an actual physical robot that does nothing but run ChatGPT software and read the prompts out loud. I hope it looks like a parrot, just to drive the point home.

        1. UnCivilServant*

          That would be funny. Does it have to sound like a parrot? I have difficulty understanding what they say.

          1. Your Former Password Resetter*

            Sounds appropriate for ChatGPT, which is also often confusing because it doesn’t actually know what language is.

      3. Drew J.*

        Isn’t that… kind of how human brains work, though? Taking a bunch of appropriate words and phrases and that we’ve remembered from past experience and using them because we think the context fits?

        Like, I only know what ‘How are you?’ means because I’ve heard it used in context thousands of times and tried it out a lot myself. In contrast, I don’t really understand ‘Ça va?’ means (culturally) because I don’t have much experience in French speaking countries. But I can still parrot ‘Ça va?’ ‘Oi, ça va.’ because I’m reasonably confident that’ll go down ok in most contexts.

        Aren’t we all just playing the odds that ‘Good morning!’ won’t be responded to with ‘No. I’m having a shit morning’. Why would we care anyway? Because we’d feel sad, right? Because our brains would give us a chemical ‘negative token’. What’s the difference between that and an electronic token, if we’re all made up of sub-atomic particles anyway?

        I’m yet to hear an answer that doesn’t boil down to ‘Mumble mumble human soul’, and hey, I can totally respect that. But if people want to bring religion in to ‘why they feel superior to electronic intelligence’ let’s bring that out in the open, so we all understand the conversation that’s actually taking place.

        1. Katie Impact*

          Are you familiar with Emily Bender’s work? She’s a professor of computational linguistics who goes into some detail about the differences between how humans and LLMs process language, and why the answer to your initial question is no.

          1. Drew J.*

            I’m of the opinion that Emily Bender doesn’t go deep enough into the biological processes of human learning. She’s doing philosophy. Take her octopus example: she assumes that any human would know what a bear or a stick is, and that therefore makes them different to an LLM. But if you raise a baby in a submarine and don’t teach it about bears or sticks it wouldn’t be any more helpful than the octopus in her analogy.

            She says LLMs are different to humans because LLMs are trained without external inputs. But that’s not a real reason, because a human being who was raised in an empty white box with nothing but words on a screen to learn from would have the same problems.

            I’m not asking stupid shit like ~Does an LLM have a mind?~ I’m asking ‘Do LLMs really learn in a radically different way to the way we learn?’ ‘Would thinking about “How would I explain this to a human who was raised in a box” help me get a better output?’. Sure, humans have extra stuff like serotonin and dopamine adding complexity, but that doesn’t really affect the way I’d learn when to say ‘Ça va.’, which is just ‘Have a bunch of experiences seeing it used in context and let your brain average those experiences out in the background for you’.

            If LLMs DO learn in a similar way, it makes them far easier to understand. Humans lie and make stuff up if they’ve been told ‘I don’t know’ is not a permissible answer. Humans struggle to give useful and accurate tech support over the phone when they can’t see Grandma’s screen and she’s just saying stuff like ‘I clicked the box and now it’s gone’. What box? What’s gone?… Have you tried turning it off and on again?

            Whether or not they ARE different, LLMs were specifically designed to store inputs in the same way as a human brain, by updating clusters of individual neurons. It makes sense that they’d LEARN in the same way as us. Whether anyone thinks that means they actually ARE like us on some level comes back to religion and philosophy again. But analogously, if you want to argue that moving a membrane to vibrate air to make sound isn’t a concept that’s shared between human speech and a set of speakers you’re going to have a hard time trying to listen to Spotify.

            1. JB*

              What point are you actually getting at?

              You keep saying what you’re asking is whether LLMs learn, mechanically, the same way humans do, and that you are not concerned with whether they have a conscious mind.

              But then you say that the mechanics of learning will determine whether or not they are lying. That’s not true, and I feel like you must know it’s not true. Regardless of under what circumstances humans learn that it is beneficial to lie, the act of lying in and of itself requires intention – that’s why repeating misinformation that one believes to be true is not lying.

              LLMs are not able to understand the concept of truth versus untruth and they are unable to lie because they do not possess intention, which is a prerequisite for lying. So what are you actually getting at here?

          2. Drew J.*

            Ok, here’s an analogy: from my viewpoint it’s kind of like somebody’s come out with a new type of delivery/item transportation robot. It’s been intentionally designed to have legs so it can fit in around work spaces designed for humans. People are getting excited because it does a great job in warehouses, hospitals, etc.

            But then people start complaining ‘I tried to get it to take my parcel to our other site across the river and it didn’t make it! This technology sucks!’

            I’m trying to say ‘Well of course it does, it was designed to walk like a human, not to fly like a standard drone. You’ll need to give it a boat to get it across the water or it will keep sinking.’

            And the response I get is ‘Stop anthropomorphising the robot! It’s not a human! Humans need boats. Robots shouldn’t need any assistance with that sort of stuff. Drones don’t. Wow, listen to me talking about the robot needing assistance, now I’m anthropomorphisng the thing. Pah! Clearly it just doesn’t work very well.’

            To which my response is ‘Wait, we designed this type of robot to mimic the way human bodies function in this particular area, why are we now trying to pretend we didn’t? It’s not like it’s a perfect one-to-one biological analogy, but we can’t pretend the imitation isn’t deeply baked in to the design.’

            Now replace ‘legs’ with ‘network of neurons’. You don’t have to believe it’s human to understand you’ll get the best results from thinking about it like one (when it makes sense to do so).

            Some free advice if anyone’s still reading – I get better and more detailed answers from ChatGPT when I say please and thank you. I don’t do it because I think it has ~feelings~, I do it because it’s trained on a vast dataset of human interactions and humans are statistically more helpful when people are polite to them. When I’m terse, it matches context. When I’m polite and grateful (in a genuine tone, it has enough statistics to detect sarcasm and empty flattery), it gives more detailed assistance (very helpful when I’m debugging complex code). Maybe y’all struggle with getting anything useful out of ChatGPT because you’re talking to it like jerks…

            1. aqua*

              “You don’t have to believe it’s human to understand you’ll get the best results from thinking about it like one (when it makes sense to do so).”
              Thinking about ChatGPT as a human is actually a great way to get terrible results from using it because the ways it fails are superficially similar but caused by fundementally different mechanisms.

            2. JB*

              Nobody struggles to get “anything useful” out of ChatGPT. Saying that the technology is inherently limited by the fact that it cannot understand the difference between reality and fiction is not saying that it inherently sucks. It’s recognizing the appropriate use for the technology.

    2. MigraineMonth*

      That isn’t a mistake, that’s working as intended!

      We taught an electronic parrot to talk . The fact that we expect the parrot to understand what it’s saying instead of just regurgitating words that it’s heard strung together before is the real mistake.

      1. T.N.H*

        I understand why hallucinations happen but it’s still a mistake the way OP means it. As in, if your robot parrot said something incorrect, presumably she’d be furious.

        1. Kella*

          If someone uses ChatGPT as a tool to say correct things, *they* are making a mistake. ChatGPT isn’t programmed to be correct or factually accurate.

          1. Drew J.*

            A hammer is a useful tool, but if you’re using a hammer when you need a screwdriver that’s either your fault or the fault of the person who told you a hammer would be useful in your context. It doesn’t somehow make the hammer a poor tool.

        2. MigraineMonth*

          Very true, I just feel the need to shout “It’s all beautiful nonsense!” from the rooftops every time I hear of someone trying to use it for yet another inappropriate purpose. There aren’t that many applications for a tool that writes grammatically correct but likely inaccurate information; spam and political speech writing, maybe?

          The absolute worst I heard is helping doctors write their notes faster: not only are there symptoms being made up, there’s just enough pattern recognition in the tool to be racist as well! They said they’d fix the “hallucinations” issue in the next release, as if it were a bug instead of a foundational problem with the technology.

          1. Drew J.*

            I find it extremely useful for all sorts of things. But I don’t just ask it to invent from scratch – I tell it the facts FIRST, then I ask it to put those facts into an email or a meeting agenda for me. If you think confidently spouting BS to save face when you don’t know the answer is just a ChatGPT thing you haven’t spoken to a teenager lately.

            Now I’ve used it for a while it’s pretty good at giving me a different email for Jane, Paula and Steve from the one set of facts. So Jane gets her email that’s straight to the point, Paula gets all the polite ‘fluff’ and Steve gets the slightly simpler phrasing because English is his second language. Yes I did need to train it with some sample emails first, but the upfront time cost was worth it.

            It’s also great at things like ‘Please write me some python code to generate a PDF of a basic meeting minutes template’.

            ChatGPT like an intern. It knows how to write but it doesn’t know any facts you haven’t told it. And like an intern, how useful you find it is a reflection on how good you are at clearly communicating what you want it to do.

            1. Nell*

              Ever read A Computer Called LEO: Lyons Teashops and the World’s First Office Computer by Georgina Ferry?

              “…In May 1955 the Association’s annual conference in Hastbourne
              was devoted to the subject of electronics in the office…There were chuckles when, in an introduction to the workings of a computer, one lecturer quoted an anonymous salesman as saying, ‘Forget the idea of a computer as an electronic brain … Think of it as an expensive moron, infinitely slow and infinitely stupid, which can make mistakes faster than you ever dreamed possible.’…”

            2. Give Me Strength*

              It doesn’t know how to write. It doesn’t know how to do anything. It is incapable of knowledge.

              1. I am Emily's failing memory*

                The use of “know” in this context is analogous, not literal. The same way a “file” on your computer is not the exact same literal thing as a “file” in your filing cabinet. It’s a readily understandable analogy that helped people at the dawn of the computer are, who had no prior concept of a computer file, quickly grasp what it is and what purpose/function it serves.

      2. Pita Chips*

        The fact that we expect the parrot to understand what it’s saying instead of just regurgitating words that it’s heard strung together before is the real mistake.

        I don’t think we can hit that point home hard enough.

      4. Working Class Lady*

        Migraine Month, I couldn’t think of a more accurate way to describe AI if I sat down and brainstormed.

    3. Strive to Excel*

      I’m wildly entertained by the chat thread where someone was desperately trying to get ChatGPT to figure out how many ‘R’s there were in ‘Strawberry’.

      1. Drew J.*

        I’m just bewildered by it, honestly. Are people amused because their mobile phone can’t pour them a glass of water? It’s such an advanced piece of technology, and such a simple task, after all! ChatGPT hasn’t been given memory storage capabilities to allow it to count, so of course it can’t count. There’s nowhere for that information to be stored because the algorithm on its own isn’t an entire Turing machine. Run the algorithm ON a Turing machine and give it access to memory storage and it’ll count just fine.

        I can guarantee you that ChatGPT is competently counting on command on OpenAI’s internal servers, they’re just not giving random members of the public the ability to tell ChatGPT to store specific stuff on their machines. It’s locked down. And with good reason, that’s how you get hacked.

        1. Meg Jo Beth Amy*

          Huh? It’s obvious on the face of it that my mobile phone can’t pour me a glass of water, because it doesn’t have the tools to hold a glass or turn on a faucet. ChatGPT, on the other hand, is something I’ve been told is an “artificial intelligence.” Intelligence, to me, includes the ability to count. Even if you get more specific and call it a “large language model,” well, counting is part of language. All the tech companies pushing AI aren’t saying “use these for everything except counting.” Thus, it isn’t at all obvious that ChatGPT should be unable to tell me how many Rs are in “strawberry.”

        2. aqua*

          “I can guarantee you that ChatGPT is competently counting on command on OpenAI’s internal servers”

          I can confidently guarantee you that it isn’t because that’s

          1. aqua*

            Oh cool my phone submitted this halfway through writing it. I was going to say:

            I can confidently guarantee you that it isn’t because that’s not how the technology works, that’s fundamentally not a thing large language models can do, and if you think they can you do not understand large language models.

        3. JB*

          You seem really invested for someone who apparently does not actually follow ChatGPT news.

          Yes, ChatGPT has always had in-session memory, and as of this year it can now store memory from one session to the next. Memory is not the problem here.

          I really think it might be healthy for you to step back from ChatGPT for a bit, because it genuinely sounds like you do not understand that it isn’t a conscious entity. You talk about it “lying” to “save face”. It sounds like you think that when it is instructed to count something, it is somehow understanding that instruction and is just unable to follow it and is making up an answer to compensate. That’s not how this tool works.

      2. amoeba*

        That one at least has been fixed in the new 4o version – it includes reasoning now and does get the strawberry problem right (and can also explain why). We had some people at work play around with it and they were actually quite impressed.

    4. Beany*

      Yeah, but … mistakes in management/social interactions aren’t the same as mistakes in technical work, and people have different aptitudes. It’s *possible* that OP really is error-free in their actual technical work, and I they’re not claiming anything like that for their interactions with their coworkers. I’m not saying it’s likely they’re 100% perfect at the technical stuff, just that one doesn’t logically follow from another.

      (This is different to the more recent letter from the clue-free interviewee who maintained their general perfectness, IMO.)

      1. T.N.H*

        I think that’s true. But a lack of self awareness usually pervades all aspects of one’s personality.

    1. restingbutchface*

      Please, point it out? Shame is a great way to learn, according to OP.

      I’m not asking because some of us have appalling grammar. (Me, I’m talking about me)

      1. Hlao-roo*

        I think the error is in this sentence: “That includes, deadlines, technologies, methodologies, features to be included, etc.”

        To the best of my knowledge, there shouldn’t be a comma after “includes.”

        1. Worldwalker*

          There shouldn’t. The commas are to set off items in the list, and “includes” is not such an item.

          So the LW made a mistake, and consequently should be removed from the team — or in this case, banned from ASM!

          1. Sandi*

            AAM, not ASM?
            I really don’t care and would have never mentioned it (and probably not even noticed) except… the topic of discussion made it irresistible.

            1. Airy*

              Muphry’s Law states that whenever you write something correcting an error in someone else’s writing, your correction will contain an error.

                3. Strive to Excel*

                  I don’t know – Murphy’s Law is “If it can go wrong it will”. Maybe Muphry’s Law is about the correction of errors!

                5. I Have RBF*

                  Is it bad that I’m sufficiently dyslexic that it took me several minutes to see the different spelling of Murphy/Muphry?

    2. Juicebox Hero*

      Grammatical error, too.
      “Whenever I see a mistake in anyone’s work, especially trivial ones…” Should be either mistakes to match ones, or a trivial one to match a mistake.

      Somehow, my blood is not boiling.

    3. Dust Bunny*

      Starting that sentence with “but” is not the biggest grammatical sin out there, too, but it would have been better had that sentence and the one preceding it been a single sentence.

    4. RagingADHD*

      I see two punctuation errors and three very questionable word usages and sentence structures.

      And I still don’t think LW deserves to be berated for being human and missing something. Teased a bit in service of self-awareness, yes.

  5. restingbutchface*

    Like a parent who reacts with rage when a kid makes a bad choice, you’re creating an atmosphere where it is safer to lie and cover up mistakes rather than taking ownership.

    In a tech environment, that’s how minor issues become major incidents. Plus, it sounds exhausting for everyone involved. This style of management needs to die out like, yesterday.

    1. dulcinea47*

      The kid didn’t even make a bad choice necessarily, that’s adding blame, they made a mistake, which is human and often accidental.

      1. Nightengale*

        my professional work some days is 60% explaining to people why we shouldn’t frame children’s impulsive behavior as “a choice.”

        1. Up the Down Staircase*

          Oh hey, mine too! I absolutely detest the “they made a poor choice” narrative. It’s lazy and completely absolves the speaker of having to look at other factors that may be relevant.

    3. UKDancer*

      Yes, in aviation also it’s critical to build a culture where people can raise issues and admit to mistakes without feeling blame. Many aviation disasters happened when people made mistakes but couldn’t admit to them or couldn’t challenge mistakes made by those higher up.

      1. darsynia*

        If anyone would like to see what that attitude looks like, the channel Mentour Pilot is superb and has done a lot towards championing the idea of crew resource management as something applicable to more than aviation! The healthy notion that an increased workload means a person is more likely to make mistakes and/or be under stress is such a healthy one that more folks could do with recognizing!

        Re: the LW, I would find such an atmosphere extremely hostile and wouldn’t want to stick around, and I imagine that most workers with options will feel similarly. They’re really fostering a situation where they lose workers, here.

    4. Turquoisecow*

      I hope OP never works with people who are still learning or early in their career because this sort of atmosphere would have destroyed me.

    5. Kes*

      The fact that they’re a tech lead makes this even weirder to me. Mistakes happen in tech all the time, that’s why QA is part of the process.

      1. restingbutchface*

        Right? I love mistakes, I wouldn’t have a job without them. Nothing is perfect (apart from OP).

    6. RVA Cat*

      That’s what I immediately thought, too. This is childhood trauma from (at least) verbal abuse parent or authority figure who demanded perfection *because it is impossible*.

      Anyone else wondering if the LW became the Hypercritcal Boss who destroyed their report’s whole sense of self? (Letter from Sept. 6, 2023)

    7. TM*

      Yep, 100%.

      Alison’s advice was great. I’d reiterate that no-one is perfect, even in tech, unless it is low-level no-brainer work (which should probably be automated). I have worked with nationally-acknowledged gurus in my specialty, and I have witnessed even them make (small!) mistakes.

      As more food for thought, it may help people like the LW to realise they are there for the TEAM. Or, if you think of the Swiss cheese model, the tech lead is the last layer before the team’s output is released. So, yes, they should catch errors, and then WORK WITH the team member to get it resolved (plus monitoring for patterns in such errors). That way, the TEAM’s output is “perfect” (as it can be) and the team lead is fulfilling one part of their role, coordinating and monitoring the team’s output. Rather than getting angry, they could then spend more time on an aspect they are not (probably) fulfilling: being a trusted resource for guidance on how to complete a task and a mentor for more junior team members. Supportive, encouraging, and not a source of unwarranted stress.

      And if the carrot of being more useful to the team as a whole isn’t enough, here is the stick: if you consistently act in this explosive, irrational way, you will lose your more skilled staff very soon. I’ve been in this game for 25 years – if someone blew up at me like that once over a trivial error, I might give it a pass in a high-stress environment. If that happened more than once over different, trivial issues, I’d simply walk, and take my 25 years of technical plus institutional knowledge with me. That kind of thing gets expensive to a business after a while.

    8. ThatOtherClare*

      Doing things that don’t work is just part of trying new things. If your team is making the same mistakes over and over, sure, that might be a sign of a problem, but if they’re never making a new mistake that means they’re doing exactly the same task again and again and never changing the process. You’ll never see any productivity gains that way. Plus you’re stuck if somebody external invents a new and better way of doing things.

      Take the industrial revolution as an example: sticking to hand weaving might have meant a weaver didn’t make mistakes in trying to learn how to use Cartwright’s new power looms, but it was the people who jumped into the messy and mistake-ridden learning process who made the profits. You don’t have to be an early adopter, but if you ban mistakes you’ll end up drinking tea with the Luddites – a cosy and comforting idea perhaps, but not a great decision from a business standpoint.

    9. amoeba*

      Yup, was coming here to say that. A good error culture is so, so important in safety-critical environments – because if people are afraid to admit to errors, that’s how things go really, really wrong.

  6. Ceanothus*

    I worked for a team where a senior reviewed certain work and was very similar to the LW in her tolerance for error. She was eventually moved to a very high-priority project — she is also something of a unicorn — and I took over the review for six months.

    People cried because I was professional and pleasant about it. After a couple of months, people were willing to dig into recurring issues because it didn’t involve being buried under contempt. We figured out the source of the most common errors and put together process and reference documents to prevent 95% of the problems that the senior was catching. When she returned, her reviews found significantly fewer errors (despite her conviction that we had just been screwing up constantly in her absence). ((She also got pretty upset when she found some errors and we were like “OK! Let’s update the references to reflect this!” because we shouldn’t NEED references for corner cases, we should just KNOW.))

    1. Slow Gin Lizz*

      Ugh. I’ve only just started working on a large technical team after being the sole tech person in a couple of very small non-profits, but from what I’ve learned so far, you’re SUPPOSED to have references, aren’t you??? At my new job, we document EVERYTHING. So not only was that senior a jerk, but she was doing it wrong. (Right?)

      1. Howard Bannister*

        Absolutely. People who rely on ‘just knowing’ take a perverse pride in being the only person who is always absolutely correct, but have no real interest in sharing their knowledge.

        It’s often something you observe in a certain kind of dysfunctional workplace because it also feels like job security — nobody else holds all that knowledge, so they’re irreplaceable. (and at my job we all say that no matter how good any of us are, none of us should be irreplaceable, because in the normal course of things every single one of us will need to be replaced)

        References and documentation are the backbone of an organization that allows it to plan for the future. Standing in the way of that and protesting it is shooting the whole organization in the foot!

        1. Slow Gin Lizz*

          Right! And actually, at my last several jobs where I was the only tech person, I documented EVERYTHING as I was working my last few months on the job. I only gave two weeks notice, of course, but once I started looking for a new job I started documenting what I did so those last two weeks would be less chaotic.

          At my last job, over the course of my first year, odd things would happen where my boss or grandboss would say, “Oh, yeah, this happens every year and I know to look out for it….” or whatever. I started to document those too, right in our database, because I knew other people would be filling in for me or when the grandboss announced her impending retirement I knew we were losing her institutional knowledge. Growing pains at a small org where suddenly there are four people doing the job that used to be one person’s, but the lack of documentation frustrated me so I started making note in obvious locations of all the weird things that happened every year.

          So…yeah. Document, document, document! And any boss or coworker who says you shouldn’t should be expelled.

      2. Disappointed Australien*

        If nothing else you need a “walked under a bus plan” other than the organisation giving up and disbanding. If your boss has that as their plan you need to find a better job.

        It’s a bit of a PITA these days with 2FA on a lot of things, but you absolutely need it. The pain of getting access to accounts is significant when you know all the details, but without them you’re left saying “but it’s {company owner} paying for this, you have to help me” (reader, they do not *have* to help you).

        Weirdly you can buy things like YubiKeys one at a time (physical authentication tokens). Even though the minimum sane number of YubiKeys is two.

    2. Goldenrod*

      “People cried because I was professional and pleasant about it.”

      First of all: Sad (and I’ve been there)!

      Secondly, this really illustrates why this fault-finding behavior is so nonsensical. Humans naturally make FEWER mistakes when we aren’t made to feel terrified of making mistakes. You can’t learn and grow in a totalitarian culture of fear.

    3. ThatOtherClare*

      Considering your approach reduced workflow errors and hers didn’t, I’d say the one making the biggest overall mistake there was her. Her approach was proven to be ultimately responsible for every single mistake that your approach fixed. She was making large, recurring, avoidable management mistakes with a direct cost to the company. Such people are often far more tolerant of their own mistakes, though, unfortunately.

  7. Sandi*

    There is also a logic that says fixing all the small mistakes isn’t worth the effort to attain perfection. For example I pride myself on writing well, and I’ll catch many of my mistakes. Yet I can’t spend days rereading the document to catch any others, especially since as humans we become almost blind to the mistakes we make. Instead, I send it off to a coworker who quickly reads it through and offers a few edits. My workplace would much rather that I do a pretty good job and have my coworker spend 30 minutes providing feedback, as this is much more efficient than my spending days trying to find perfection. If I’m writing code then I’ll ensure it works well, but I use it for small internal projects so I don’t spend weeks perfecting its efficiency as that would be wasteful overall.

    I feel like the LW is wasting a lot of time and effort trying to find perfection. And wasting a lot of emotional energy and relationships with coworkers!

    1. Dawbs*

      it’s rather like typing that way.

      when i was a child and learned to type on typewriters (tar pit ink on mammoth hide), we were docked 5wpm for every error (considering how long it took to correct, i probably should have been docked more. Which meant i could for 45 wpm perfectly OR i could type way faster, but “officially” still around 45 wpm once the errors were tallied in.

      Now that there’s an easy way to have my”teh” and “tge” get instantly changed back into the “the” i was trying for, even though my speed on the keys is slower (finger issues) i can really claim 55wpm.
      i could type with fewer errors, but correcting is faster than allowing my fingers down

    2. Dawn*

      Great news, major corporations releasing major external projects don’t spend any time perfecting their code’s efficiency nowadays, either.

      1. MigraineMonth*

        For the most part, they shouldn’t worry about the code’s efficiency unless it is an actual issue. Even in performance-critical code, so much time and effort is wasted (and complexity introduced) for small or non-existent efficiency boosts.

        Example 1, I worked for a company that used extremely short variable names (so “v1” instead of “averageWeight”) and instead of the null check (v1 != ‘ ‘) did an equivalent but difficult-to-read “does not follow” check (v1 !] ‘ ‘). Why did we do these things that made the code harder to read and errors more likely? Because 15 years ago, those made the code slightly more efficient before the compiler was updated and made them redundant.

        Example 2, I was given a program that requested information from an API and wrote it to a database. This process ran nightly and took about 6 hours, thanks to the optimization that used multi-threading to pull data from the API with five simultaneous processes. The only problem? 95% of the time was writing to the database, which required exclusive locks, so the processes pulled lots of data simultaneously and then waited in line to save it. I actually undid all the multithreading because it was too much of a headache to maintain for the 30 minutes saved.

        As Donald Knuth said, “We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil.”

        1. Dawn*

          I’m talking more about the AAA game companies who don’t bother optimizing their code because “they can just buy another hard drive” here.

          We used to squeeze entire operating systems onto a floppy disk and now the newest Call of Duty is 175 GB and just increases our rare earths deficit because people keep having to buy more storage space.

    3. londonedit*

      This is the case even in book publishing. There are people who LOVE to crow about finding errors in books – but the fact is, we’re all on deadlines. If the thing doesn’t go to press on press day then we’re screwed, because the one thing you can’t do is miss a publication date. Everything is already set up in terms of printing, marketing, sales, publicity.

      Of course, as part of the editorial process a manuscript goes through: structural/development edit, followed by a review by the author. Then a copy-edit, followed by a review by the author. Then a proofread and another review by the author. Then I go through it two or three further times to try to catch any remaining errors. Then the author does another review before press. Then I check everything one final time. But at some point, it has to go to press. And I’m dealing with several books all at the same time, all at different stages, so I don’t have time (and nor is it my job) to do a line-by-line read on absolutely everything. And so, yes, the odd mistake will creep in. It’s like that phrase ‘perfect is the enemy of done’. At some point the book just has to be done.

  8. ReallyBadPerson*

    How do we know this isn’t the same guy, now (finally) employed after convincing an employer of his perfection? I mean, have this LW and that one ever been seen together IRL?

  9. Priscilla Tells It Like It Is*

    Honey, your behavior is a big fat blatant obvious mistake. Let’s address that first before having meltdowns at your team members.

      1. Pastor Petty Labelle*

        People are removed from the project for one mistake! That is so messed up, I cannot even begin. This is affecting people’s professional development, their standing at work, and possible even their evaluations which affect their income. OP needs to stop affecting people that way. And bosses needs to stop allowing OP to do so.

        2. jlp*

          I mean, depending on the consequences, I might just weaponize incompetence and intentionally make a mistake to get off the team – but that’s just a random thought.

        3. owen*

          it’s also affecting the project as they are now down one resource until they get another onboard and caught up – a process that might (shock!) involve mistakes as they are learning… and now you need ANOTHER new resource….

        4. New Jack Karyn*

          I suspect that after a while, getting bounced from OP’s team would not be seen as that large a black mark.

  10. duinath*

    “The rationale in my head is always “We have ONE job and one job only, and that’s to get this done! No excuses.” “

    This is the bit that stood out the most, to me. I feel like it demonstrates such a misunderstanding of how life, and indeed most tasks at any job works, that I do see this statement in itself as an excuse.

    At most workplaces, it’s not “one job” it’s a series of things that need to be done. And even if the job itself is something incredibly simple and not complex, people have lives. People live in the world. Things can be difficult for reasons that have nothing to do with work, and things can fall through the cracks for so many different reasons.

    If you won’t work with anyone who ever faltered even slightly, you cannot work with *anyone*, not even yourself. Your mistake here is a very serious one. More serious, I would guess, than any that led to your heated conversations.

    I hope this letter gave LW the wakeup call they needed and things got better for them. This sounds like an unpleasant way to live.

    1. girlie_pop*

      This stood out to me as really weird too. Maybe if your job was to staple packets of paper together all day, every day, then yeah, you have “one job”? But basically, nobody in the professional world that I’ve ever met has “one job”. Everything we do has multiple steps and stages; we have to coordinate with other people and collate information, and even with checks and balances, mistakes are going to happen. Acting like the whole process is so simple and basic is not going to help anything.

      1. Slow Gin Lizz*

        “You had ONE JOB” is the kind of phrase I use for things that don’t work right. Like, say, a can opener that doesn’t actually open cans – what even good is it? But humans? We all have more than one job, right from the beginning. Even newborns are supposed to eat, sleep, stare at the ceiling, and other biological functions (of which there are too many to even count). Any one of these things can interfere with every other thing (butterfly effect, in other words). And while maybe OP means that their ONE JOB is to get things right, there are so many other jobs that have to be done right before that ONE JOB (get things right) can happen, which by definition means they all have more than one job.

    2. Miss Muffet*

      And what are even the consequences for the errors? Very few situations are going to be life-or-death. Some may have bigger consequences than others (say, an error that drops someone from their medical coverage right before a surgery) but can be fixed. And doesn’t software development usually come with testing before things go live? So hopefully you’re catching the errors there and that’s LITERALLY WHAT THAT PHASE IS FOR.

      1. metadata minion*

        And the situations that are life-and-death usually have checklists and more than one person signing off on the finished product, precisely because humans aren’t perfect and are going to miss things.

        1. MigraineMonth*

          Bringing this kind of attitude to managing a project with life-and-death consequences will literally kill people. You CANNOT make people afraid to admit mistakes, talk about their mistakes, or put in processes to detect similar mistakes when the stakes are high. That’s why they aspire to a “blameless culture” in safety-critical fields like medicine and aviation.

          1. ThatOtherClare*

            It should be the goal for any team really.

            ‘So, why did your bridge fall down?’
            ‘Because Timmy sucks! I hate him!’
            is not a dialogue I would accept from a teenager, let alone an adult at work.

            That’s not why the bridge fell down and you know it. Go and take some time to cool off, and when you come back we’ll all have a chat together about the different ways we could help Tim make the upstream pylons stronger.

            1. N C Kiddle*

              The interesting thing is my knee jerk response would be to be Timmy in this situation. “Because I screwed up. I’m a useless excuse for a human being and I should go and eat worms.” I can see with the application of logic that this is not a helpful response either, but as someone mentioned upthread, the wrong kind of parents will do that to you.

    3. Lana Kane*

      I know this is an older letter but I think that mindset is a reflection of a lot of “motivational” social media content that we see out there. No excuses! We all have the same hours in the day! etc And it’s easy for some to see that as all very logical and apply it to their lives in a whole-cloth way.

    4. Roland*

      Same! I’m a software engineer and I very much do not have “one job”. There’s always many things to be done.

  11. Ellis Bell*

    I feel like there’s a real disconnect on what this person’s job actually is and what they think it is. It’s phrased as though they’re not responsible for working with people, just responsible for judging the work: ” I decide whether a piece of work by any team member is acceptable” but these kind of interpersonal skills and responsibilities are considered basics and often never spelled out in job descriptions. Just because you’re not a manager doesn’t mean you don’t have to manage relationships. I would say to OP: 1) How many perfect team members who never make mistakes do you have, and are there enough of them? 2) Is the mistake easily fixed and easy to learn from? 3) This habit of writing people off because of errors is probably your mistake. How would you like someone to speak to you about fixing it? Angrily?

    1. Paint N Drip*

      I totally agree on the disconnect. It seems like they’re a tech person who refuses to incorporate non-tech things into their internal job description, even though it seems obvious that soft skills are part of this reviewing task and thus the job is more than just the “cut and dry” coding or whatever. Yet again, soft skills being undervalued!!

    2. MigraineMonth*

      How many people is OP booting off these projects, and are they getting replaced with new people? If so, I can’t imagine these projects aren’t blowing their deadlines; getting people up to speed on a project is a huge time sink.

      Not to mention the issue with too many cooks having contributed to the soufflé, even if half of them have been thrown out of the kitchen: “Well, Bill designed it as an instantiable class, but his replacement Blair moved most of the logic to a static utility, then their replacement Alexis extracted a helper class that calls a JS library, except her replacement Sue is uncomfortable with JS so she encapsulated it…”

  12. Boss Scaggs*

    I hate to say this but your team probably really dislikes you if this is how you treat them. I dont’ have any actual advice other than perhaps talk to a therapist. It’s not healthy for anyone to continue to behave this way

    1. Anon for this one*

      Seconding this.

      (Also, LW, for the sake of stigma reduction, OCD does not cause people to act like this and it’s a misconception that it does. However, your deep-seated resentment and entrenched beliefs about how the world should work are definitely worth talking about in therapy.)

      1. Cat Tree*

        Thank you for clarifying that. I have OCD and it doesn’t make me an asshole. It’s also worth pointing out that most adults with OCD recognize that the disorder is affecting them, at least to some extent.

        1. ThatOtherClare*

          Yep, OCD affects what comes into your head, not what comes out of your mouth. That’s all on you bud. Even when people with Tourette’s or ADHD say something hurtful, the next sentence out of their mouth is ‘Oh no, not again. Sorry!’, not #DealWithIt.

      2. Higgs Bison*

        Yeah, it would be closer to OCPD if it were anything, and it’s irrelevant either way. The key takeaway is that it’s doing OP, their team, and their company a disservice.

  13. RagingADHD*

    The entire reason for having a review and approval step in the process is to catch errors. The presumption that there will *always* be errors that need catching is built in to the LW’s job function.

    If it were abnormal and unacceptable to ever make any errors in the work, there would be no need for LW to have a job at all, and everyone would just be pushing code or publishing documents, or whatever, based on their own inherent perfection.

    LW is not only completely out of line and counterproductive for reacting this way, they are highly illogical in thinking their own job should not need to exist.

    1. Paint N Drip*

      That’s a good point. I wonder if the OP’s perspective on the job is more ‘I don’t make mistakes so they choose me to catch yours’ rather than ‘this process is inherently at risk of mistakes so it’s industry standard to have a system in place to catch them’

      1. RagingADHD*

        It goes to LW’s understanding of how things “should” be. As humans, we react strongly when our sense of what should / should not be is violated, and the further our expectations are out of line with reality, the more we overreact or react inappropriately to ordinary situations.

        If the LW believes the number of errors in the work submitted for review should be zero, their expectations are out of line with the existence of their job. The number of error in the finished product should be zero (or as close to zero as humanly possible), but the work submitted for review is not finished until they have done their job.

        Their job is to catch and correct errors, not berate people.

    2. Slow Gin Lizz*

      LW should thank their coworkers for making mistakes that LW can fix. In one of my first jobs, I was taught that this is job security.

  14. TeapotNinja*

    There is only one reason to get upset at someone for a mistake. And that’s if it was done deliberately. The person knows it was going to break something and did it anyway. Even then there’s a professional way of dealing with that.

    1. Loreli*

      If it was done deliberately it wasn’t a mistake. It was a bad decision. Mistakes are inadvertent.

      1. Aye, Aye, Ai*

        Even setting aside the semantics, I disagree with the original statement. If the worker is making *enough* mistakes, or doing nothing to fix them or avoid them, that is surely a reason to “get upset.”

        1. UnCivilServant*

          Getting upset won’t fix anything.

          Collect documentation, try to politely and calmly address the errors with the employee. If they continue to resist implementation of processes to reduce errors, bring it to the manager (as the LW isn’t in a hire/fire position.)

        2. MigraineMonth*

          Agreeing with UnCivilServant: as a manager, you should never be upset or angry over poor performance. You need to go through the process of letting the employee know that it needs to improve in order to stay employed, and then follow through with letting them go if they don’t improve, but that shouldn’t come from a place of anger or frustration.

          Believe me, it’s hard enough hearing you don’t measure up and are going to lose your livelihood without having to deal with your manager’s emotions on top of that.

    2. Cat Tree*

      Honestly even then there is some nuance. My company (manufacturing) is big on human error prevention. When someone knowingly and deliberately does something wrong, it can be an organizational benefit violation, personal benefit violation, or reckless violation. The first two are potentially fixable to a certain extent. Only reckless violations warrant getting upset, and those are the most rare.

        1. Cat Tree*

          An organizational benefit violation could be not replacing a part as frequently as required to save the organization money (and typically with a history of seeing that part last longer so not expecting a bad outcome). A personal benefit violation could be skipping a step to save time so you can leave work early (and again, typically with some prior experience of not viewing that step as necessary so not expecting a bad outcome). A reckless violation would be an angry person intentionally destroying equipment.

          1. Cat Tree*

            Actually I thought of a good example. I work in sterile manufacturing where operators have to don sterile gowning to go into the suite, and press their fingers and suits to agar plates when leaving to prove that they remained sterile. There was an event in a different department where a guy peed in a jar while in the sterile suite. I actually don’t know how he physically managed it because the suits are all one piece without a fly, but anyway. It was discovered a part of a larger investigation for a batch that failed sterility testing.

            On the surface it seems so obvious that you don’t pee near a sterile product. But getting out of the suite and then back in would take at least half an hour. We staff just like any other company, so there probably wasn’t a person to take over for him. But we gotta meet production schedules so a long break didn’t work. This was an organizational benefit violation because he thought he could save the batch. Honestly he might have done it multiple times before but nobody ever knew about it until it actually contaminated the product. So maybe he figured it wouldn’t do any harm.

            On one hand, what he did was gross and outrageous. On the other hand, the company knows that employees have biological needs but give them grief for it while demanding high productivity and low staffing costs.

            This is of course a very extreme example, but a memorable one.

  15. Buffalo*

    Talk about a silly nitpick, but it’s bugging me – Alison’s response includes:
    (Which, as people are pointing out in the comment section, is a mistake in itself! So there’s some irony there.)

    Wouldn’t her response have been written before anyone in the comments section was pointing out anything?

    1. Juicebox Hero*

      Sometimes she adds to her response, like if someone adds relevant information or points out why her advice won’t work.

    2. duinath*

      She sometimes edits her advice after the comment section gets going. Best guess, if you find the original it’ll be noted there as an edit, but here I don’t think that’s necessary since it’s noted this is an older letter.

    3. there-are-roaches-in-my-pie*

      This letter was originally published in 2017, so she probably added it in once this sentiment gained enough traction in the comments. I’m assuming since she’s just rerunning the letter, she’s not going to edit it out.

      1. Good Enough For Government Work*

        I think Alison probably has more and better things on her mind this week than editing her old responses to letters.

    1. Elle*

      For real. I know I’m not the only technical editor rolling their eyes.

      Also: ART, is that you?

    2. Grammatik*

      I read OP’s letter three times and am not seeing the run-on sentence, although she should put a comma before “etc.”

      In general, though, yes, run-on sentences are absolutely a mistake, and it is not a subjective matter to be decided by someone “critical enough.” If you’re overlooking run-on sentences, you shouldn’t be an editor. I also have no problem screening job applicants based on grammar.

  17. CommanderBanana*

    If you worked for me, regardless of how ‘perfect’ your work was from a technical standpoint, I would fire you.

    1. Pastor Petty Labelle*

      Certainly remove them from reviewing other people’s work and never ever promote them to a position of managing people.

    2. Slow Gin Lizz*

      I really want an update from this LW. I want to know how long they were working in this job, how long they’ve had the power to remove ppl from projects for one mistake, and if they were able to keep this job. I figure either LW took Alison’s and commenters’ advice and changed their attitude (which: yay!) or they unfortunately were let go from this position (which: boo).

      1. Captain dddd-cccc-ddWdd*

        The 3rd and imo more likely possibility is that LW left this job of their own accord (without changing their attitude) with the reasoning that it’s a company cultural issue they couldn’t fix, “the company” tolerates too many mistakes from people, doesn’t have high standards or hold people accountable, etc. In my mind they’ve carried this attitude with them to interviews and gone into great length responding to questions like dealing with poor performance, rectifying a mistake, conflict with someone, etc.

  18. Naomi*

    “We have one job and one job only!” is a very misleading way to think of this. In many jobs (especially in tech!) people have to carry out complex tasks with many components. If there’s a mistake in the work because someone had 50 steps to do and got 48 of them right, it’s not fair, compassionate, or productive to treat that as a complete failure.

    This is why the entire field of QA testing exists: because software is complex, and it’s extremely unlikely to be perfect on the first try. The best you can do is check carefully for mistakes and correct them before the product goes out.

    1. Double A*

      Yeah most jobs that can be described as “You had one job only” have been offloaded to automation at this point.

    2. Ellis Bell*

      It’s ironic because OP’s job is not one thing it also involves dealing with people, like in most roles, and it would serve OP well if they started branching out more. But these kinds of catch phrases, like “we have one job” tend to be picked up from somewhere. Id’ be interested to know if the OP has ever been in previous environments that were punitive, intolerant and so narrow in perspective that they miss the point.

  19. Czech Mate*

    In defense of OP: we don’t know what kind of work these folks are doing or what the stakes are for making mistakes. If they’re writing launch codes for nuclear missiles and they’re regularly making the same errors, then yes, that’s pretty serious.

    But! If that’s the case, then safeguards need to be put into place to catch errors/typos when they happen (à la NASA launch status check). If the work is that serious, you need to assume that you will always need multiple sets of eyes on something to double, triple, quadruple check that everything is a go.

    1. MigraineMonth*

      I said this above, but this behavior would *actually kill people* in a safety-critical environment. Medicine and aviation are two that are famous for trying to create “blameless” reporting processes specifically so that people can make and admit their mistakes without getting fired. It’s also important that everyone admit that *they* might have made a mistake and be open to listening to corrections from even the most junior member of their team, which I suspect LW also isn’t doing.

      If a patient is seizing, is it better to have a nurse who immediately admits he might have administered the wrong dose, or one who clams up because he doesn’t want to lose his job? If the plane is heading straight towards a mountain, is it better to have a pilot who has a track record of listening to his copilot, or one who is so arrogant or abusive that his copilot defers to him until they crash into the mountain? (If the latter seems farfetched… it unfortunately happened.)

      Yes, if someone’s performance is bad because they’re making a lot of avoidable mistakes, you should flag the mistakes and notify their manager (the person *actually* in charge of managing their performance). That’s nothing like getting enraged and berating everyone about every trivial mistake.

      1. anon24*

        Yup. I work in EMS and we try to create a culture where we can admit mistakes, because it’s far less likely to do long term harm when you can go “oh shit, I think I may have done the thing” than to hide it and wonder why the patient is dying. You’ll still face disciplinary action, but speaking up is encouraged.

        I once made an admittedly irresponsible mistake that had nothing to do with patient care but did end up damaging equipment and I called my supervisor right away and admitted what I did, why I did it, and took full responsibility. I got written up, but at a lower level than what company policy dictated, because I took full responsibility right away and didn’t try to avoid it.

        New people sometimes get told all the stories of our fuck-ups not to scare them or be like “hey, we’re incompetent,” but to show them that no one is perfect and also please please learn from everyone else’s mistakes and we will tell you all of ours so you don’t have to make those same mistakes because they hurt so badly to make and if you can learn the lesson I learned without having to do what I did then we are all better for it. Share the knowledge, not the pain.

    2. biobotb*

      The OP doesn’t describe pulling people off projects for making repeated mistakes. They say even a trivial mistake will trigger removal of the person who made it.

    3. Ellis Bell*

      I can’t think of a single environment where being punitive would make things safer. It would just make cover ups more likely. As you say, the only safe way to operate is with the assumption that mistakes happen and therefore lots of checkpoints are needed to catch them.

      1. N C Kiddle*

        I think some people are imagining two extremes, either getting really mad at every tiny error, or shrugging and saying errors are never a big deal, and assuming that since we think the former is bad, we must think the latter is the way to go. Which, no. There’s a middle ground called understanding that some errors are a big deal and putting procedures in place to catch them without getting mad.

    4. RagingADHD*

      There is a safeguard. LW is supposed to be the safeguard.

      They are getting mad at other people because they have to do their job (flagging errors). Which suggests maybe they shouldn’t have a job that they resent doing.

      I can’t imagine how they have anyone left on any projects to do any work, if they pull everyone who submits a single error off a project. It is incredibly dysfunctional.

  20. Lisa*

    I work on highly-regulated, life-sustaining medical products, where mistakes could literally kill someone. Even we do not expect people to never make mistakes, because human beings make mistakes. Instead we put in place systems and processes that make mistakes less likely and also catch mistakes so we can fix them. The LW’s expectations are unreasonable, and they need to focus on what to do to mitigate mistakes, not get angry when they inevitably happen.

  21. Aye, Aye, Ai*

    On the first issue, people are going to make mistakes because you work with humans, not robots, and humans make mistakes.

    I’m not disagreeing with the above, but…look at the longshoremen’s strike for its implications.

    1. Orv*

      Longshoreman aren’t threatened with being replaced by robots because they made too many mistakes, though. It’s because robots would be cheaper.

    2. Ginger Cat Lady*

      Your comment makes ZERO sense. Longshoreman mistakes are not an issue in the strike.

  22. Jax*

    This was a tech lead writing in, and I’m curious whether the phrase “trivial mistakes” in that context is kind of the opposite of how most of us think of it? (What makes me think it is that OP said it is especially the trivial mistakes that are upsetting.) By which I mean: One wrong character or space in coding — what seems like a trivial mistake — could keep everyone searching for hours or days to find and correct it and torpedo deadlines in major projects and divert resources from focusing on substantive areas? It might be a big deal is what I mean and shouldn’t be happening even once.

    Not that it makes a difference in recommendations to the OP on seeking help in controlling emotional reactions in this context.

    1. MigraineMonth*

      I don’t think a tech lead would describe a mistake that torpedoed deadlines as a “trivial mistake.”

      Even if that was the case, OP is profoundly misunderstanding their role as a reviewer. They are supposed to look for mistakes, flag them to be fixed, and only approve when all the mistakes they can find have been fixed. That is a completely normal step of software development everywhere I’ve worked, and nowhere did that person fly off the handle because they… had to do their job?

      OP isn’t supposed to be managing anyone’s performance; if there is a pattern of avoidable mistakes by a certain team member, they can flag that to the person’s manager.

    2. ThatOtherClare*

      If your deadlines rely on none of your human beings ever accidentally dropping a semicolon you’re going to have a bad time. There’s a reason why QA Engineer is a whole entire job.

      I once spoke to fascinating man who was one of the QAs for the Bank of England’s first computer. He told me the programmers would get mad because he used to break their code by doing things like typing his name in a field that was supposed to be ‘numbers only’ (yes, that long ago, we’ve come a long way). Apparently they told him to stop doing it(!) and he refused. I remember his indignation as he asked me ‘Why would they expect that people would only ever do what they’re supposed to do? The stakes were too high for that. We had to assume that people would get it wrong and safeguard ourselves accordingly.’

  23. Erelen*

    Never mind that the emotional and heated reactions probably cause some anxiety in even the most confident people, thus causing MORE mistakes to happen.

  24. Insert Pun Here*

    I work in a job/industry where mistakes are Not Acceptable and a high degree of polish for anything public-facing is the norm, and the way you prevent mistakes is by having everything reviewed by multiple people. Expecting individuals to never make mistakes is unrealistic.

  25. MakingBiscuits*

    Tech VP here. I *guarantee* that a tech lead like this wouldn’t last long in any dev shop I’ve ever been associated with. His people would hate him, and (being and knowing nerds) would actively work to get him fired/removed, which would almost certainly be accommodated pretty quickly by somebody in his chain of command.

    As Allison noted, there are two options here. Therapy or unemployment.

  26. Caz*

    I work in IT, and frequently have “one job to get right”. “One job” is almost never one *step*, and we get it right by having up to four different levels of review – all asking essentially the same question (“does it work?”) but all looking from different perspectives. Despite all that, mistakes occasionally slip through! They might be minor, such as typos; they might be more significant, and might even impact health outcomes for individuals (I work in healthcare). We do our best with all of them, we sort what we can sort whenever we come across it, and work with services to put things right when they raise them. 100% accuracy is absolutely not an expectation of our team, and my manager will happily regale us with tales of times that he cocked up! This LW would benefit from some time in my team…but with his attitude, he’d never be hired.

  27. Lisa*

    I hope this person has been able to get help for their issues. Sounds like a painful way to live!

  28. Fluffy Fish*

    Soft skills are as important if not more important than technical skills.

    So as critical as OP is of others, they are actually failing 50%+ of their job.

  29. Coverage Associate*

    In the developed world, especially in tech, there aren’t many functions that theoretically could be automated that have not already been automated. I’m struck by how much discretion OP has while having the attitude that other team members only have 1 job.

    I admittedly don’t know a lot about tech, but in all the jobs I know, things like deadlines and methodology are not black and white. If they were, it wouldn’t be a person’s job to decide them. There would be nothing to decide.

    And in well run places, deciders get input on things like methodology and internal deadlines (especially if the decider isn’t approving leave time).

    My long winded point is, in addition to everything emotional already pointed out, perhaps OP is also calling “mistakes” what are actually differences of opinion.

    1. Roland*

      As a software developer, it’s definitely very rare that I see developers other than interns doing things that are really, truly, no-opinion-involved mistakes. There’s better and worse ways to do things, technically and project-eise workflow-wise. But it’s almost all subjective.

  30. Honoria Lucasta*

    I mean, props to OP for at least being self-aware enough to recognize that this is bad, and for being able to describe their own reaction clearly. I could imagine many worse and less honest ways of writing in about the same issue — you know, the LWs who say “My team hates me because I have Standards, and sure I raise my voice a little bit but they deserve it.”

    At least this OP asks “how bad is it?”, admits that they “sort of know it’s not good”, and asks for general advice given the whole set of circumstances (rather than specific advice on how to keep doing the same things without experiencing consequences).

    ISTM that this is what taking a first step toward positive change can look like!

    2. Mad Scientist*

      I completely agree and thank you for this. A lot of commenters are acting like the LW was unaware of the issue or unwilling to change, when it’s actually the opposite. Sometimes we know our behavior isn’t ideal and want to improve but simply don’t know how. Kudos to the LW for recognizing this and trying to change!

      To be honest, I struggle with feeling similar to the LW sometimes. Not to the same extent, and I don’t really act on it (nor do I have the power to), but I empathize with feeling frustrated by incompetence (or what you perceive as incompetence) and feeling unsure whether or not your expectations are unreasonable. I try understand that mistakes happen, we’re all human and I make mistakes too, but when I see frequently repeated or easily avoidable mistakes, that just comes across as sloppiness to me and it really tests my patience! Again, I don’t necessarily act on my impatience or disappointment, but I do feel it sometimes.

      All this is to say, the LW’s feelings are understandable (to me), and what’s important is how they act on those feelings. I hope they’ve been able to improve since they wrote this letter!

  31. Strive to Excel*

    Apart from all the other excellent points Alison and the commenters have made, it’s common that having people fix their own mistakes is a bad use of time! Humans tend to start not noticing mistakes in our own work after a while, because our brain ‘fills in’ with what should be there. So it takes more time and energy. You can also have a situation like writing where the person creating the content is not the person checking for grammar and spelling mistakes, because those are different skillsets. If you are ok with them being done by separate people, you broaden your ability to get good people.

  32. Jake Purralta*

    A lot of their team will end up so anxious due to their behaviour they’ll end up making more mistakes and it will end up in a vicious circle and the employee will lose all their confidence. I’ve seen this happen numerous times over the years. Once to me, I left a job after 2 months as it was awful, it took me a year to get my confidence back.

  33. Lab Snep*

    I worked at a lab where someone ONLY ever communicated with me when I made a mistake, and then was really awful about it.

    Even minor errors that had been caught before going anywhere and immediately rectified were the worst thing ever.

    I was blamed for mistakes I didn’t make. When I pointed that out I was never apologized to.

    It was one of the most demoralizing experiences ever.

  34. WaffleWizard*

    I think it’s worth a mention that this is a great way to stifle innovation too. Who on earth would want to try new things you’re not super familiar with if you’re going to have to deal with a manager having an emotional meltdown if it isn’t perfect the first time?

  35. Boof*

    Frankly even machines make “mistakes” / errors / mechanical failures / etc
    It’s odd that they are MORE angry about trivial mistakes too – like why does their emotions go up the lower the stakes are?

  36. Goldenrod*

    I remember this one well, because when it was first posted, I saved it so I could re-read it (which I’ve done, many times).

    I just love Alison’s response. I had a boss like this, and Alison so perfectly articulates and analyzes all the problems with this type of behavior.

  37. RVA Cat*

    Is LW is Jane the Hyper-Critical Boss from September 2023? Though even she seemed to be more contemptuous than angry.

  38. Larry, I'm on Duck Tales*

    Seems like the way this person deals with others is a bigger mistake than the routine ones made by the mere mortals they work with.

  39. Heffalump*

    It was well known at Apple Computer that when you showed Steve Jobs something you’d done, you didn’t give it your best shot the first time around. He’d tear it to pieces no matter what. He once said, “You’ve baked a lovely cake, but you’ve used dog shit for frosting.”

    1. Goldenrod*

      Yes, and from what I’ve heard about him, he was not actually a good leader after until he left and came back with an improved attitude.

  40. xl*

    I’m an air traffic controller, and this person’s demeanor would not cut it here.

    If that’s the case, then I think that it’s safe to say this person is overreacting and taking themselves way too seriously.

  41. WillowSunstar*

    The other thing is, don’t forget Perimenopause can also lead to mistakes made by people who didn’t used to make mistakes. There are actual physical changes happening to our brains, but the US really has no protections for this stuff. All we can do is try to adapt andhope we don’t get fired.

    1. I wear my sunglasses at night*

      I’ll be honest, I’m not really seeing what this has to do with the letter? It’s also giving “misogyny but in a sneaky way.”
      First it’s, “oh those women, they can’t anything done when they’re on their periods—so hormonal!!”
      Then it’s “oh those women—why should we expect decent work out of them when they’re distracted by getting a husband, or having a baby, or taking care of the babies they already have???”
      And I guess now it’s “Tsk, well of course you can’t expect a WOMAN to produce good work—-she’s going through THE CHANGE!!! Speak softly and don’t look her right in the eyes or else might she has a mood swing and freak out on you for no reason!”

      I have enough to worry about with going thru perimenopause right now—I don’t need the sift bigotry of low expectations to add to it, thanks but no thanks.

  43. DramaQ*

    I could swear this is my manager. Any error I make no matter how small is held against me and my performance till the end of time.

    I’m regularly threatened with not being allowed to develop or do parts of my job I enjoy if I cannot become as close to perfection as possible.

    I challenged it asking out of the 200+ submissions we get per month how many am I making mistakes on? 5 out of 200 compared to 150 out of 200 is quite a difference for example.

    I was told any errors are unacceptable.

    We have a review system but was told that is not what it is for. Then why do we have it?

    I end up making more mistakes because I’m so stressed trying to be perfect. I’m burnt out and deeply resentful.

    As soon I find something else I am GONE.

    The LW is the worst type of micromanager and they will drive out all but those who have no other options. The people left will circle the wagon and start lying about/hiding mistakes to protect each other.

    And they never will figure out they are the problem.

  44. New Jack Karyn*

    A bigger problem here is OP’s manager. She’s not reining him in! She’s the one with the power to check his behavior. It sounds like the manager assigns people to teams, and she is ceding that control to OP by allowing him to kick people off his teams for any mistake.

    If one of the teammates had written in, we’d say that yes, this guy is being a jerk–but his manager is also a major part of the problem by her inaction.

    1. Ellis Bell*

      If OP is having “sometimes heated” conversations defending their position with their manager, I would say that being reined in was on the horizon. In fact, I assumed that was what prompted the letter to Alison, because OP knows their behaviour is not approved of.

  45. nnn*

    I wonder if OP could channel that energy into mistake-proofing their team’s processes, using their technical expertise and attention to detail to figure out what needs to be adjusted so mistakes can’t slip through. Maybe they’ll find that intellectually satisfying, maybe that will scratch the itch in their brain that arises when mistakes occur.

  46. Spinner of Light*

    Very Important Point: A bad temper is NOT a strength – it’s a weakness! It shows that you are NOT in control of yourself, that your first reaction to being balked or thwarted or disappointed is to fly into a rage and that you do so without regard as to how you’re coming across to others.

    So what does that sound like? That’s right – a toddler throwing a temper tantrum! If you don’t respect a tantrum-throwing toddler, why on earth should your colleagues respect YOU if that’s how you behave? If you have authority over some of them, they may FEAR you, but that is NOT the same as respect! (Confusing fear and respect is another hallmark of immaturity, by the way – the two are very different.) Rest assured that the targets of your temper are less than impressed…and less than respectful.

    Your colleagues may react in many ways to your tantrums, but some of the more common reactions are to cover up their mistakes in an effort not to anger you, quietly get a little of their own back by subtly sabotaging you, or going to YOUR supervisor with documentation of all the times you’ve blown up at them. If you continue to throw tantrums, be prepared for any or all of the above to happen.

    If you don’t like these scenarios, it’s within your power to change how you respond to normal human error. And that is the wonderful difference between you and a two year old! Toddlers CAN’T yet control themselves. An adult can – and is expected to!

  47. MCMonkeybean*

    If you yell at people for making mistakes, then no one will tell you when they’ve made a mistake, and the final result will be that your projects are full of way more mistakes.

  48. Double Shelix*

    Oh, i feel like i have this guy for a customer.

    Long story short, i compose and edit long technical documents. The one in question was 250 ish pages, 340 steps, and in the 35000-40000 word range. The customer sends cover sheets of desired edits with rationale, so when they complained about the error rate in my medium-but-not-early draft, i sat down and counted. Per word, my error rate was 0.02 %. Percent. So 0.0002 errors per word.

    I was livid and told boss, boss’s boss, anyone relevant who would listen that i challenged anyone in my department, or an adjacent department, to beat that “score”. Spoiler: that was far fewer typos and incorrect step references than usual, especially for a document that size.

    I still work with him two years later, and he acts like he trusts me, but deep in my heart i know the relationship will never be solid again.

Comments are closed.