I believe that to prevent Juliuses from proliferating one needs to stop making up for their mistakes.
Generally I have found that allowing things to fail is a great way to communicate a reality to management.
If we’re indeed facing a future where AI is dreamt up to fix things that it can’t, we’re going to have to illustrate its limitations my allowing this to result in some pain. The earlier the better, because then the pain is relatively small.
In a perfect world, sure, but in reality I don’t think this is good advice. When you’re assigned to fix something, don’t say “I can’t, the previous person built it too shitty.” The only thing anyone hears is that you cant. The previous person has already been promoted. Go ahead and leave and let them “learn their lesson”, just don’t be surprised when they don’t and do fine.
I think the lesson might be that raw engineering competence just isn’t that critical to the success or failure of a business. Find somewhere you like and don’t expect success, or find somewhere successful and don’t expect to like it. There’s no market evolutionary pressure to make places with both, so they’re very rare. Definitely don’t expect any correlation between success and coworker competence.
“I can’t, the previous person built it too shitty.”
I agree, this doesn’t work, because this is a claim someone without an engineering background can’t verify.
In my experience, if there is a clear pattern of lack of competence, what does work is:
Make it clear what the roles and responsibilities are in a team, making sure management has signed off on it. In my experience, an effort to better define or document roles and responsibilities is always welcomed by people in management positions.
If one of the team members can’t meet their responsibilities, don’t silently do their work for them, instead allow them and the team to face / learn from the consequences.
It’s easy to mistake this for a passive aggressive move, I believe however that this is in fact treating everyone as adults; for example, one lesson learned can be that the role someone aspires is not a good fit. In a different role they might shine.
Having a team where it is allowed to fail is worth a lot. And again, the earlier you are allowed to fail, the lower the cost and the higher your efficiency in the long run.
I agree that is often the case, but how could they? They don’t have the skillset to look at the code and quality is really hard to measure (e.g. reported bugs is a very meager measure of quality, since only reproducible and isolatable bugs are reported by users).
In the end it’s a matter of trust, and I think it’s up to both parties to develop that trust. For instance, I personally believe that it’s up to the engineers, where possible, to develop ways to make quality quantifiable, rather than just saying ‘believe me, the code is bad’.
Absolutely. For example, I just merged a PR to one of our public repositories that adds a form of git diff --stats to the visual programming language Max/MSP that my department works with, which gives insight in the quality delta of a refactoring effort. This has worked great to illustrate the value of an engineering effort to management.
I have to keep reminding myself that our economic system is built to reward Julius. They’re not at fault for simply being the most fit to survive in the environment we created.
I feel like the system, such as it is, is made up of people, and I can be cranky, simultaneously, at both the aggregate at large and the individuals that come within melee range.
This inspires me to write my own blog post, I’m aware that the author has something clear to say about AI tools, but it would be an immeasurable catharsis for me.
I worked with a man so deeply charismatic (as described) that his failures glided off his back like water off a goose. He caused (and continues to cause) significant harm to the company - I left because I was tired of doing my own job and trying to make sure the entire company continued to exist as well.
After I left, he was promoted. He reports to essentially nobody and blames everyone else when things fail- which seems to be working for him. As expected, the company is suffering- extremely badly: nobody in management will point at him though. It’s very impressive.
among the ones who are self-aware, what do they think of themselves?
I speculate that they see some very limited self-awareness. They probably think of themselves as “middle-of-the-pack” fiftieth-percentile types. I see extreme outliers on the other end of the spectrum from Julius and I think of myself as “decent, but not quite as genius as ‘Xyz’ outlier.” Perhaps Julius thinks the same about their colleagues. “They’re suited to tasks A, B, C and I excel at ‘formatting, presentation, and meeting management’. It’s good that we can decompose the work and each do what we’re able.”
It’s hard for any of us to see our shortcomings, IMO. It’s not a bug, it’s a feature. Once you start to dwell on “man, I am really a ball-dropper” you probably start to experience what we’d diagnose as “depression.” So we have some cognitive dissonance and we do “wave away concerns” to protect our ego. This allows us to perform to the best we’re able.
Regarding 3., they just never think about it. It’s not something that occurs to them, if they’re qualified or not, or their level of competency, or if they’re a net positive or negative. That aspect of metacognition isn’t important to them.
I too have worked with at least one Julius. It’s comforting to know I’m not the only one. I’ve always wondered - does Julius do this kind of thing intentionally as a survival technique, or is it more like natural selection?
A little bit of both. Julius has got to eat. He’s also not qualified to do anything else. But he’s hardworking and the system really wants to reward him for what he does do.
I once had the pleasure of joining a team staffed by a manager (technically useful variety) and a young Julius. Absolutely lovely guy, could barely follow tutorials. The manager had to let him go and in the meantime Young Julius’ career seems to have blossomed.
I can tell you that this Young Julius was not deliberately gaming the system. He genuinely wanted to be technically good - in fact he wouldn’t take my advice to work less hard on work and try doing some of his own projects. I’m certain that he’s found success doing what he’s actually good at.
One thing I’ve realized over the years is that it’s not necessarily a great injustice that Juliuses get promoted, just that upper management might be something for which they’re actually good at. It takes less technical skill but demands a high level of communication, personal marketing, confidence, adaptability, things which are still important skills nonetheless.
I disagree, they are an absolute disaster in management, at any level.
A mid-level manager needs to be able to communicate in both directions, up and down the hierarchy. They need to be able to communicate concerns from the factory floor to senior management and communicate high-level objectives down. They don’t need to understand the details of how everything is done, but they need to be able to understand what is possible and how difficult it is. And, most importantly, they need to know when to say ‘I don’t know, I’ll have to consult my tech lead before answering that question’ rather than promising something impossible. A Julius absolutely doesn’t know how to do that, they will keep promising impossible things. Often, they’ll be able to get promoted and move to a different business unit in a reorg just after that on the basis of what they promised, and the disaster when their successor fails to deliver on the promise is blamed on someone else, so you may also lose a good manager as well as have a failed product.
In senior management, they are easily bamboozled by mid-level Juliuses (Julii?), who will tell them amazing and infeasible things are possible. They will then create strategies around things that their companies are unable to deliver. They don’t know anything about the way that the product is really built, because they never did it and other people fixed their messes, so they won’t prioritise any of the things that are core to their business and will let quality drop until customers notice and move to a competitor in droves.
I saw a lot of them at Microsoft, it sounds as if Boeing also has more than their fair share.
Overall I don’t disagree, they are definitely proving their worth for upper management track.
However I’ve seen more than one Julius in my career who very much stayed technical. I feel like Joel Spolsky’s Architecture Astronauts post is about exactly this, and I have personally witnessed Juliuses (Julipodes?) get promoted into Principal or Architect roles and just ruin eng orgs.
We had one director once, don’t want to say Julius because that is my name, but similar. We saw him buying vodka in the morning before coming to the office. Always smooth, always liked. But man…
What I’m learning from TFA, the comments here, and the comments on HN, is that I should have named my son Julius so he would guaranteed be successful in life.
I have to say I’m a little surprised and disappointed that no one here came to stick up for Julius and say that C does have a virtual (fine, an abstract) machine … ;^)
I’m not gonna lie, he got me in the first half. At first I thought the article was yet another rant against a strawman from a cranky, dysfunctional senior programmer convinced software development is mostly a technical problem. I didn’t expect it to be an allegory.
In another life I was a lawyer at a large law firm. Large law firms mostly need Julius. Most of the work is within Julius’ grasp. Promotion, especially to the partnership requires getting a bunch of clients who view you as their lawyer, and obviously this requires developing Julius-nature if you don’t have it.
The only truly dangerous people in a law firm are the ones who are completely Julius but think they’re technical.
I believe that to prevent Juliuses from proliferating one needs to stop making up for their mistakes.
Generally I have found that allowing things to fail is a great way to communicate a reality to management.
If we’re indeed facing a future where AI is dreamt up to fix things that it can’t, we’re going to have to illustrate its limitations my allowing this to result in some pain. The earlier the better, because then the pain is relatively small.
In a perfect world, sure, but in reality I don’t think this is good advice. When you’re assigned to fix something, don’t say “I can’t, the previous person built it too shitty.” The only thing anyone hears is that you cant. The previous person has already been promoted. Go ahead and leave and let them “learn their lesson”, just don’t be surprised when they don’t and do fine.
I think the lesson might be that raw engineering competence just isn’t that critical to the success or failure of a business. Find somewhere you like and don’t expect success, or find somewhere successful and don’t expect to like it. There’s no market evolutionary pressure to make places with both, so they’re very rare. Definitely don’t expect any correlation between success and coworker competence.
I agree, this doesn’t work, because this is a claim someone without an engineering background can’t verify.
In my experience, if there is a clear pattern of lack of competence, what does work is:
It’s easy to mistake this for a passive aggressive move, I believe however that this is in fact treating everyone as adults; for example, one lesson learned can be that the role someone aspires is not a good fit. In a different role they might shine.
Having a team where it is allowed to fail is worth a lot. And again, the earlier you are allowed to fail, the lower the cost and the higher your efficiency in the long run.
This assumes management can tell the difference between “done” and “done well” and that it cares. In my experience, it doesn’t, and it doesn’t.
I agree that is often the case, but how could they? They don’t have the skillset to look at the code and quality is really hard to measure (e.g. reported bugs is a very meager measure of quality, since only reproducible and isolatable bugs are reported by users).
In the end it’s a matter of trust, and I think it’s up to both parties to develop that trust. For instance, I personally believe that it’s up to the engineers, where possible, to develop ways to make quality quantifiable, rather than just saying ‘believe me, the code is bad’.
Have you ever tried actually doing this?
Absolutely. For example, I just merged a PR to one of our public repositories that adds a form of
git diff --statsto the visual programming language Max/MSP that my department works with, which gives insight in the quality delta of a refactoring effort. This has worked great to illustrate the value of an engineering effort to management.The problem I think is that organizations are just as good at identifying Juliuses as they are at attributing failure to Juliuses.
In all likelihood the failure would blow back on some innocent party.
This is usually because you mostly have Juliuses above you. Somewhat hyperbolic but: only Juliuses hires Julius.
I have to keep reminding myself that our economic system is built to reward Julius. They’re not at fault for simply being the most fit to survive in the environment we created.
I feel like the system, such as it is, is made up of people, and I can be cranky, simultaneously, at both the aggregate at large and the individuals that come within melee range.
I wonder if this is a org chart inversion of analytical ability. People below the middle of the chart know what the business does better than the top.
The exceptions to this at large orgs is in my experience so few they dont have impact
This inspires me to write my own blog post, I’m aware that the author has something clear to say about AI tools, but it would be an immeasurable catharsis for me.
I worked with a man so deeply charismatic (as described) that his failures glided off his back like water off a goose. He caused (and continues to cause) significant harm to the company - I left because I was tired of doing my own job and trying to make sure the entire company continued to exist as well.
After I left, he was promoted. He reports to essentially nobody and blames everyone else when things fail- which seems to be working for him. As expected, the company is suffering- extremely badly: nobody in management will point at him though. It’s very impressive.
The questions I have are:
I speculate that they see some very limited self-awareness. They probably think of themselves as “middle-of-the-pack” fiftieth-percentile types. I see extreme outliers on the other end of the spectrum from Julius and I think of myself as “decent, but not quite as genius as ‘Xyz’ outlier.” Perhaps Julius thinks the same about their colleagues. “They’re suited to tasks A, B, C and I excel at ‘formatting, presentation, and meeting management’. It’s good that we can decompose the work and each do what we’re able.”
It’s hard for any of us to see our shortcomings, IMO. It’s not a bug, it’s a feature. Once you start to dwell on “man, I am really a ball-dropper” you probably start to experience what we’d diagnose as “depression.” So we have some cognitive dissonance and we do “wave away concerns” to protect our ego. This allows us to perform to the best we’re able.
Regarding 3., they just never think about it. It’s not something that occurs to them, if they’re qualified or not, or their level of competency, or if they’re a net positive or negative. That aspect of metacognition isn’t important to them.
I too have worked with at least one Julius. It’s comforting to know I’m not the only one. I’ve always wondered - does Julius do this kind of thing intentionally as a survival technique, or is it more like natural selection?
A little bit of both. Julius has got to eat. He’s also not qualified to do anything else. But he’s hardworking and the system really wants to reward him for what he does do.
I once had the pleasure of joining a team staffed by a manager (technically useful variety) and a young Julius. Absolutely lovely guy, could barely follow tutorials. The manager had to let him go and in the meantime Young Julius’ career seems to have blossomed.
I can tell you that this Young Julius was not deliberately gaming the system. He genuinely wanted to be technically good - in fact he wouldn’t take my advice to work less hard on work and try doing some of his own projects. I’m certain that he’s found success doing what he’s actually good at.
One thing I’ve realized over the years is that it’s not necessarily a great injustice that Juliuses get promoted, just that upper management might be something for which they’re actually good at. It takes less technical skill but demands a high level of communication, personal marketing, confidence, adaptability, things which are still important skills nonetheless.
I disagree, they are an absolute disaster in management, at any level.
A mid-level manager needs to be able to communicate in both directions, up and down the hierarchy. They need to be able to communicate concerns from the factory floor to senior management and communicate high-level objectives down. They don’t need to understand the details of how everything is done, but they need to be able to understand what is possible and how difficult it is. And, most importantly, they need to know when to say ‘I don’t know, I’ll have to consult my tech lead before answering that question’ rather than promising something impossible. A Julius absolutely doesn’t know how to do that, they will keep promising impossible things. Often, they’ll be able to get promoted and move to a different business unit in a reorg just after that on the basis of what they promised, and the disaster when their successor fails to deliver on the promise is blamed on someone else, so you may also lose a good manager as well as have a failed product.
In senior management, they are easily bamboozled by mid-level Juliuses (Julii?), who will tell them amazing and infeasible things are possible. They will then create strategies around things that their companies are unable to deliver. They don’t know anything about the way that the product is really built, because they never did it and other people fixed their messes, so they won’t prioritise any of the things that are core to their business and will let quality drop until customers notice and move to a competitor in droves.
I saw a lot of them at Microsoft, it sounds as if Boeing also has more than their fair share.
Julii is correct since it’s a Latin name 😁
Overall I don’t disagree, they are definitely proving their worth for upper management track.
However I’ve seen more than one Julius in my career who very much stayed technical. I feel like Joel Spolsky’s Architecture Astronauts post is about exactly this, and I have personally witnessed Juliuses (Julipodes?) get promoted into Principal or Architect roles and just ruin eng orgs.
We had one director once, don’t want to say Julius because that is my name, but similar. We saw him buying vodka in the morning before coming to the office. Always smooth, always liked. But man…
What I’m learning from TFA, the comments here, and the comments on HN, is that I should have named my son Julius so he would guaranteed be successful in life.
Or denied entry to every position because the hiring managers all read this article.
Always fun to see an indirect allusion to The Importance Of Being Earnest.
Actually the plural is Julii.
J/k. I just find it fascinating how English stacks “contradictory” case endings on top of each other in unadapted borrowings.
I have to say I’m a little surprised and disappointed that no one here came to stick up for Julius and say that C does have a virtual (fine, an abstract) machine … ;^)
I’m not gonna lie, he got me in the first half. At first I thought the article was yet another rant against a strawman from a cranky, dysfunctional senior programmer convinced software development is mostly a technical problem. I didn’t expect it to be an allegory.
In another life I was a lawyer at a large law firm. Large law firms mostly need Julius. Most of the work is within Julius’ grasp. Promotion, especially to the partnership requires getting a bunch of clients who view you as their lawyer, and obviously this requires developing Julius-nature if you don’t have it.
The only truly dangerous people in a law firm are the ones who are completely Julius but think they’re technical.