Hacker News Reader

Comment by dougmwne | Hacker News Reader

As a technical person, you have a natural moat around your castle. Technical work is difficult, complex and requires years of training and experience.

Business and management work is... not nearly as complex. But the people doing that work have the same highly complex brains as you do. They can handle the same deep complexity as you. So with all this extra mental space, they add layers and layers of complexity to seemingly simple business processes. This complexity builds the moat protecting their turf from others. It creates the irreplaceability that you get for free.

Most of this complexity has no clear business purpose (bureaucracy). Some of this complexity is fiction or invented languages (bullshit). It is all designed to exclude outsiders and prevent anyone not on the inside from being able to tell what is going on or easily get anything done.

View on HN

> Business and management work is... not nearly as complex. But the people doing that work have the same highly complex brains as you do. They can handle the same deep complexity as you. So with all this extra mental space, they add layers and layers of complexity to seemingly simple business processes. This complexity builds the moat protecting their turf from others. It creates the irreplaceability that you get for free.

This is one of those fictions that engineers love to believe about management because it imbues a sense of superiority. Who doesn't like feeling like their job is better or more important than others'?

Having been an IC, middle manager, upper manager, and back to IC a few times, I can tell you this is a false notion. Business and management are different problems, but they're not necessarily "easier". As an engineer-minded person by nature, I honestly found the technical problems easier to handle than a lot of the management problems I had to deal with. Managing a computer program is generally easier than managing people. People are hard.

There is “complex” and there is “hard” and they aren’t necessarily the same thing. A good strategy is rarely complex and often very straightforward, but consistently executing to it is hard. A technical problem may be complex, but it might be easy to actually implement against and especially easy to make sure it stays implemented.

Many management and business problems fall into the “straightforward to understand but hard” category. The apparent “unnecessary”complexity around these problems is sometimes self important fluff, by far, far more often it represents individual or organizational learning about what is actually necessary to solve these “simple, but hard” problems.

I agree that people are hard, but they are hard specifically because of what I said, they need to protect their turf and conquer new turf. Management problems are difficult because of politics, and at the root of politics is self and tribe interest. Having management meetings be a bit impenetrable is a symptom of politics.

I disagree and think you have a cynical take. The root of politics is resource distribution. Because this is complex, humans do a job that is a mix of self-interest, Compromise, and (hopefully) some vision.

When it works, we don’t even see it. When it fails, we do what our brains love to do: over extrapolate incidents to patterns and lock it into our confirmation bias.

Given the current state of the governance of the US, maybe you can forgive some cynicism.

The trick is that not all of the complexity is protective politics -- some of the complexity is carefully created diplomatic language to keep together a coalition of cranky managers who managed to agree on _this very specific version of The Plan_.

And unless you're savvy, it's difficult to tell which part is which, thus the problem OP is having.

Computers are hard, but do precisely what they are instructed to do even (unless actually defective). Correctly translating what you want to instructions the computer understands is hard and complex.

Humans are hard, and cannot be forced to do anything no matter the instructions. Individuals may be coerced or motivated or bargained with in specific ways. Correctly translating what you want to instructions the human understands is also hard and complex.

The real answer is probably somewhere in the middle. Using ubiquitous language from DDD would be a godsend to bridge the gap of understanding between tech and business, but every single time I’ve pitched it to the business side they have absolutely balked, which makes me suspicious of their motives.

This is seductively true-sounding but I don't think it's true.

Imagine a business person saying, "Oh, programmers protect their work by making it needlessly complex. Every few years they invent new tools and insist that the old ones are broken, so that they can re-implement the same features over and over. Today they say it's vitally important to switch to the new thing, jquery or docker or react or whatever it is this month, and five years from now they'll say it's just as important to get rid of it. That's how they make sure they'll always have a job."

That's the mirror image of what you're saying. Perhaps some of those "layers and layers" really are useless but most of them are probably to satisfy auditors or something similarly opaque.

That businessperson would be correct too.

> Imagine a business person saying, "Oh, programmers protect their work by making it needlessly complex. Every few years they invent new tools and insist that the old ones are broken, so that they can re-implement the same features over and over. Today they say it's vitally important to switch to the new thing, jquery or docker or react or whatever it is this month, and five years from now they'll say it's just as important to get rid of it. That's how they make sure they'll always have a job."

Would they be so wrong? The state of the art is advancing, and these re-implementations can be huge improvements...but aren't always. The tools/techniques often seem to be used outside the domain where they're important (e.g. Big Data tooling used on data sets that can be processed on a nice laptop) or are so short-lived the switching cost is never reclaimed (e.g. always using the trendiest Javascript framework/tooling). This hypothetical business person certainly wouldn't be alone in wondering if the reason isn't sometimes job security, resume building, and/or to make a boring problem interesting rather than an attempt to find the most straightforward way of solving the business problem.

I can easily see the same applying to these business ideas. They sometimes make sense but often (usually?) don't.

Sure, but I don't believe anyone advocated switching to jquery (or, a few years later, getting rid of it) consciously thinking it was a useless thing to do; and similarly I don't believe any biz person introduces a process thinking, "haha this will help widen the bureaucratic moat around my team".

The truth is, a lot of those people spend a lot of time trying to streamline processes and remove bureaucracy, and if they didn't, it'd be even worse.

I returned to an office that I'd worked at, in the exact same role, 15 years earlier.

It's interesting to see the differences. The work is the same, the same people do it, but there's much more of it, there's slightly better tech, and they have correspondingly less time to do the work. There's much more middle management, but ostensibly the systems are all doing very close to the same thing.

The thing is they've been through 3 complete reorganisations of the ~1200 people. If they'd left it exactly the same structurally then it would seem to be in the same shape as it is now.

They've been through 2 renamings, 3 rebrandings, they definitely did nothing; the placement of the business is entirely unrelated to the brand image but each new CEO came from outside of the company seemingly used it, rebranding, to make the organisation "their own".

One of the experiences early on in my second tenure was a group doing a Myers-Briggs type activity, just like they would have been 15 years earlier, just with a slight difference twist. People there in the interim described a third-style of the same process they'd been through.

Maybe you're not wrong.

> each new CEO came from outside of the company seemingly used it, rebranding, to make the organisation "their own"

This is the hacker equivalent of wanting to do a greenfield refactor of the code base. There's an erroneous assumption all the problems that exist within the current code base will be avoided when rebuilding. These efforts are occasionally successful, but mostly end up being just sound and fury with little measurable results.

It is true. What your imaginary businessperson is saying is also true.

It's bullshit jobs all the way down.

You're absolutely correct. Of course this phenomenon applies to exactly the framework and platform musical chairs that the industry plays. Politics is human and exists in all areas of life. Many things on the technical side have absolute reason to exist, because it would be impossible to build a complex system without them. Some things are BS. On the flip side, some things on the management side are absolutely essential to the profitable functioning of a company, but there is also a huge amount of room for BS to thrive. It's possible for profitable companies to exist basically due to the law of large numbers, but this is not possible on the technical side unless you are training ML models.

One way to really understand the organisational BS phenomenon is to study the really old books on administration. Read China's Han Feizi, Legalism; India's Chanakyan Arthashastra; Perhaps even Machiavelli (less recommended - since he was not really a hands-on leader dealing with issues of the day actively). In these works you will see naked desires of individuals expressed clearly.

The men who established bureaucracies truly saved humanity from collapse, from utter chaos; by establishing laws and regulations, they restrained the animal within man, even without having any technology. I consider the establishers and enforcers of law the least appreciated in human history. It is a humongous task, to take up utter entropy and convert it into a semblance of civilisation, where the animal in man is kept within limits.

These things were invented long back, 2000+ years ago, and the formations have stood the test of time. They are still in use, the same ruthlessness is used to control human impulses. However, at a shallow level, our language and our "aspirations" have moved on. So we like to speak in more flowery language, hide the animal in man in public, and have forgotten why were the laws and regulations put in place by our ancestors.

I believe human regulation, through means other than bureaucracy is on the way. This alone can help humans get rid of the need to BS. With technology there is a way to combine both education/training & laws/reguations. In place education/training allows for "kind correction" while automated punishment will ensure "harsh correction". Both are needed for successful existence of human civilisation.

Just got Arthashastra. It’s amazing. Han Feizi is next on my list.

I know it's not what you are going for, but this reads like the opening to a dystopian novel - maximum control over our impulses leads to maximum coordination but at what cost? Without some disorder there is only incremental advancement and life is boring. This isn't universally good.

incremental advancement > blowing ourselves up in exuberance

tolerable life > interesting life that ends in blow up

good enough > good

I feel "universal good" is too high a standard (pure fantasy even) for our species. I feel our species may disappear from the planet or with the planet at any arbitrary moment in the future; we seem not to recognise how fragile our existence truly is.

It feels like there’s a middle way here, between these extremes, that would be the best to follow :)

Thanks for the recommendation, comments like this is exactly why I come to HN.

How does this trope still get repeated so often? It has been debunked time and time again: strategy and management are infinitely more complex than almost all technical work. Assuming a group is incompetent just because you don't understand what they're doing (even when they dumb things down to just a few buzzwords for the desperate hope that talking like a third grader will enable you to follow along), is such an obvious folly that it should at least raise some red flags in your head. Even if you feel this way you should at least consider how repeating it is of no help to anyone.

"strategy and management are infinitely more complex than almost all technical work"

Lets not get ahead of ourselves now. It's folly to write off management, marketing etc's jobs as pointless, but its equally foolish to believe what you've just espoused.

There is only so much complexity that a single human can grasp, and that is why we specialise within organisations.

> Most of this complexity has no clear business purpose (bureaucracy)

Bureaucracy has a very clear purpose. In business, politics, or otherwise.

Bureaucracy makes change difficult and slow.

Bureaucracy should be applied when you have a working solution to your problems and you don't want anyone (new devs, new managers, external forces) to change it.

Bureaucracy is often prematurely applied, in an attempt to "bring order." This cannot work. Bureaucracy applied to chaos cements the chaos into place.

Yes! Bureaucracy is helpfull. Sometime, bureaucracy prevents more bureaucracy. The issue is that it is hard to understand how or why it was set up, and to differentiate bad bureaucracy from good bureaucracy. Both look the same and feel the same. Both can be simplified. Only, when you simplify bad bureaucracy, you can get whatever. If you simplify good bureaucracy, you will definitely end up worse, at least for a time.

This is absolute poetry. It's like the Zhuangzi meets Dilbert.

The bureaucracy that can be changed is not the true bureaucracy.

Sure, not all bureaucracy is misused, it can have a good function in governance. But it absolutely can be used to extend and protect power as well. It is a tool, plain and simple.

Bureaucracy is organization. If you build something complex, you need a lot of humans to do it. But the humans need organizing principles, common goals, divisions of labor and leadership.

To not understand that, and like some of the comments above to place technical work of an IC above that is ridiculous folly. Skyscrapers and bridges and big complex things like software requires human organizational structures which we view as 'bureaucracy'.

Businesses are quite complex. A lot of bureaucracy starts as an attempt to organize and simplify the problem space.

There are simply too many decisions for one person to make and informal processes fall apart at scale.

This is also where things take on a life of their own. Without strong guidance, you end up with the rules becoming more important than reality.

Disagree on business and management work not being complex. It often involves a lot more uncertainty and you often don’t have enough data.

Management is definitely not easier. It’s dealing with abstractions around intractable problems that aren’t well defined as well as people at the same time. IC work feels easy after this which is why so many managers go back to being ICs all the time.

Reply on news.ycombinator.com