Hacker News Reader

Comment by kstenerud | Hacker News Reader

I've been in situations like this before. The end result is one giant popularity contest, with the beautiful people clique getting the highest ratings, and everyone else getting varying degrees of shit based on how unpopular they are and how much intrigue is currently going on.

They didn't even have to actively collude; humans come with very sensitive popularity meters built-in, so they instinctively know where to place others, and it pretty much governs itself.

Just imagine if all companies were run like your high school social hierarchy, except that this also determined whether you kept your job or not.

View on HN

>with the beautiful people clique getting the highest ratings,

I don't think you're wrong, but it's also a cynical interpretation.

Another way to look at it is that people who other people want to work with get high ratings. Is that bad? From a company culture standpoint, I'm not sure it is. In fact, I often see posts and comments on these very boards about the toxicity surrounding "really good programmers" who are jerks, but tolerated. That's not a great outcome, either.

Where is the middle ground?

> Another way to look at it is that people who other people want to work with get high ratings. Is that bad? From a company culture standpoint, I'm not sure it is.

Depending on who is it that people do and don’t want to work with, we might call it “discrimination against a protected class”, which is at least bad enough that it’s illegal in many countries.

It comes from life experience. I once made the mistake of contradicting a powerful person at my company at a party, and it embarrassed him. Shortly thereafter, people who had been friendly with me became noticeably cooler. My peer reviews took a nosedive and I was let go in the next round of layoffs.

That's how things work in the real world, but this kind of review system makes it MUCH easier to sabotage others when you have the clout to do so, because there's no manager to protect them from the bullshit anymore. It's basically inviting these sorts of shenanigans rather than forcing saboteurs to do a lot of extra work to get people fired.

>I once made the mistake of contradicting a powerful person at my company at a party, and it embarrassed him. Shortly thereafter, people who had been friendly with me became noticeably cooler. My peer reviews took a nosedive and I was let go in the next round of layoffs.

Obviously I have no idea the circumstances of your experiences here, so don't take this personally, this is hypothetical: but what if "you" actually were the asshole here? Wouldn't you be saying the same thing? Maybe this incident made people realize that "you" weren't a good person to work with?

If you were on the other side, isn't that a good outcome?

Maybe, but somehow I doubt that the company would lay off an exec for insulting a peon at a party.

You don't "suddenly" notice that someone is an asshole. You just know it, and avoid them as much as you can.

If someone is friendly with you for months, years, and then suddenly becomes cold, that's not because you're an asshole - it's because of some incident that either they witnessed, or was gossiped to them.

When everyone around you reacts this way almost at once, you know for sure that it was gossip.

If I'd spoken to this person along the lines of "OMG WTF are you saying dude that's so stupid" rather than "Actually I think person X had a good point about trying this approach - it would solve X and Y and Z all at once", I'd be inclined to believe that it was just me.

Person X was actually the CTO, so I thought it was a safe topic, but I'd misunderstood the shadow power structure. It turned out the guy I was speaking to had more power (at least over me anyway), and was opposed to the CTO.

How would the dot system make it easier to hurt you?

The dot system would probably *help" here, because a sudden drop in ratings right after the incident would merit an investigation.

In theory they could, yes, but in practice people would rather just get rid of trouble rather than investigate it because it's far less work.

Almost any feedback system can also be used a weapon.

> Another way to look at it is that people who other people want to work with get high ratings. Is that bad?

It is if there is no objectivity to it. It wouldn't be the first time I get a bad review because I do things slightly different.

I mean, by the same logic, you could also say it's also a cynical interpretation of the high school clique...

Another way to look at [high school cliques] is that people who other people want to be around get high [social] ratings. Is that bad?

From experience: yes, it is. The fact that this criticism can be reframed in a way that hides the problem, that something we've all experienced becomes invisible, shows that it is a problem. There is a kind of problem with group social dynamics that can't be easily ironed out; the real experience of existing high school cliques is proof of this. Unsurprisingly, this same problem that comes up in high schools appears in workplaces also.

This sounds like you've never had to go through performance reviews.

Externalities, Nash equilibria, and fairness are good keywords to start understanding why the “natural state of affairs” can suck for the majority of people.

> humans come with very sensitive popularity meters built-in, so they instinctively know where to place others, and it pretty much governs itself.

Great insight! It's like an ouija board. It feels like you're making independent decisions and a gestalt outcome "appears", but it's much more social than we perceive.

> The end result is one giant popularity contest, with the beautiful people clique getting the highest ratings, and everyone else getting varying degrees of shit based on how unpopular they are and how much intrigue is currently going on.

Welcome to life. Human social hierarchies have always worked like this. The only thing this dot collector is doing is presenting it explicitly in a way even people with poor social acumen can easily see.

Interestingly male social spaces are actually less focused on appearance than female ones. The quarterback’s popularity doesn’t rest on whether or not he’s pretty.

Whether or not the social dynamics of big established companies and high risk startups vary due to this is interesting to consider.

So, because it’s a base human urge, we shouldn’t ever try to improve on it or do better?

Sounds very defeatist to me. Plenty of workplaces try a lot harder: setting specific goals to achieve in order to get promotions rather than relying on interpersonal relationships, for example.

> Interestingly male social spaces are actually less focused on appearance than female ones.

I think it’s telling that your example here is high school. IMO the focus is always on who has the most power. In high school, yes, the pretty girls have power, because of the way the boys treat them. I can’t say I’ve seen the same dynamic play out in adult workspaces.

> So, because it’s a base human urge, we shouldn’t ever try to improve on it or do better? Sounds very defeatist to me.

Improve and do better for whom? Those of us with adequate social acumen can deal with human nature just fine.

It is observably true though that low status persons often fantasize about somehow overturn the social order such that when it all shakes out they come out on top. It’s so common that it’s a well-known and frankly overworked trope.

The world has seen plenty of revolutions and other social disruptions and every single time those with a good grasp of human nature come out on top. You can fight reality if you wish, but you can never beat it.

> Plenty of workplaces try a lot harder: setting specific goals to achieve in order to get promotions rather than relying on interpersonal relationships, for example.

And it’s telling how reliably chumps fall for that story. The process is observably no more than a fig leaf covering office politics except perhaps at the levels where nobody with influence gives a fig.

You may not think male appearance matters, but it does. Height, hair color, voice timbre, skin complexion/health all increase a man's popularity.

I’m well aware. “Less focused on” doesn’t mean “has no effect.”

We had something like that at one of my jobs. The most senior engineers there were myself and the founder, and everyone else was either fresh out of college or fresh out of high school. The founder had to spend a lot of time on business stuff so most of the things that required a senior engineer were done by me.

Most of what we did was Windows and Mac utility software, usually difficult or tricky stuff that messed heavily with system internals. The way we wrote most of this is that I would write the drivers or file system hooks or other kernel code we needed, and would also write the user mode code to support that, but not write a GUI for it or an installer. My user mode support code would either be an application that would provide a local network interface that a GUI could talk to, or a DLL that could be used by a GUI. For development I'd included a simple terminal emulator that just provided a glass TTY in a window in which I'd have a command line interface so I could develop and test my stuff without needing a GUI.

The founder went on a metrics kick and decided that the way the quarterly profit sharing should work is that everyone would be given a list of all the employees, and would have to assign to each either 2 points, 1 point, or 0 points, indicating whether they thought that employee deserved twice the average amount of profit sharing, just the average amount, or no profit sharing. You had to assign 2 points to 1/4 or the employees, 1 point to 1/2 the employees, and 0 points to 1/4 of the employees.

The points for each employee would be totaled, and the 1/4 with the highest points would get double profit sharing, the 1/4 with the lowest points would get nothing, and the rest would get regular profit sharing. Furthermore, the people with the 8 highest point totals would be on a committee for the next quarter that would largely choose the direction of the company (with the founder being able to override them).

His expectation was that I would be the top vote getter every quarter. After all, every product we currently sold and every product in development was at its core my code.

What actually happened was that everyone in engineering gave me 2 points. Everyone outside engineering gave me 0 or 1. Take the people in testing. They dealt with alpha, beta, and release candidates. They almost never dealt with me. So to them the people doing all the work on the products were the junior engineers who wrote the GUIs and those were who got the 2 point votes. Same with support. If they interacted with an engineer most of the time it was one of the junior engineers because most of the issues were with the GUI. If the underlying problem was with my code support would probably still talk to the junior engineer who write the GUI, who would be the one to determine it was actually a lower level problem and bring it up with me.

Another person who scored low who clearly should have been in the top quarter was our head IT guy. In many ways he was even less visible than me to most employees. There were also some people in operations that were critical to our success and would be hard to replace--surely that means they deserved double shares if anyone did.

This was exactly what I told the founder was going to happen and he had to quickly scrap that system.

This is a fantastic example in how incentive design can go awry! Would love to hear any other stories like this one (from you or anyone) to learn from.

Me too

Learning through others' mistakes instead of ones own :-)

Also, interesting to read

At least they didn't double down!

You don't have to imagine it, it's already the way it is. A company social hierarchy is just a little more forgiving than a high school.

Well, maybe not _all_ companies...but most of them? Yeah , actually. High school sounds about right.

no, you imagine a well-functioning high school with that colorful story.

Reply on news.ycombinator.com