jimberlage

I worked at a company that emulated Valve’s hyper-flat structure on their engineering team, with 1 manager having 50 direct reports. That’s as close to a management-less structure as I can think of, since your manager can’t attend meetings or do 1:1s anymore.

It’s great at the beginning. We started with a team of mostly self-motivated people and the lack of upward review made technical decision-making easy.

Eventually, you hire someone who is not self motivated. Also, some existing people get wise to the fact that no one will check up on them, and read Reddit for half the day.

About 4 months in, every team had 1 person like that. They had to work around them - one team can’t ever get designs cause the designer is checked out, one team’s backend work takes 1.5x as long as everyone else.

People say things to the underperformers, but there’s no teeth to anything, no one is anyone’s manager, so it’s just suggestions. They get ignored. Resentment builds into each individual team’s culture. Deadlines start slipping.

1 year in, non technical leadership is fed up. They don’t see benefits from the flat structure, and hire a new CTO and new middle management layer.

The new managers come in briefed with “the team is lazy.” The underperformers get pushed out, and have trouble finding work because their skills have absolutely atrophied. Any remaining high performers are permanently tarred with the reputation of the org from the flat structure days, and get micromanaged. To the new managers, they are kids who will misbehave the second they aren’t watched (which, in fairness, is kinda what happened at the organizational level when they weren’t watched.)

Sone good middle management providing timely oversight and feedback could’ve avoided the whole situation.

Out of curiosity, the OP’s language is “quality issues”, not “quality issue.” Why did you assume there wasn’t already a pattern of behavior implied there?

> “I’ve noticed quality issues in your code recently that’s resulted in some rollbacks. Can we talk about how we can address that?”

That’s the first step in fixing the quality issues, not an end state. Reports don’t know what they don’t know, so step 1 is to get their ideas on how to fix quality issues.

> We were managing 47 Kubernetes clusters across three cloud providers.

What a doozy of a start to this article. How do you even reach this point?

My understanding is that Internet cafes are/were more common in India.

So I guess Perl is a gateway drug for the APL family of languages now?

Sounds like there were three major stakeholders in your work at Company B.

One thing that is always on the table - if you see a person picking up valuable work and they don't have a ticket for it - you as a manager can create that ticket.

Now you may need to coach the person on how to do that themselves (can we make the ticket making process more lightweight? Can we make a heuristic like just put story points for the time you've already spent on it plus a buffer for after-the-fact work?)

But managers who really want documentation and truly think people are doing underappreciated work can always make it themselves.

Honestly, this doesn't strike me as a bad layoff announcement, with one caveat.

I’m writing to let you all know that after careful consideration, we've decided to reduce our global workforce by approximately 20% or 528 Dropboxers.

Should be replaced with

I’m writing to let you all know that after careful consideration, we've decided to reduce our global workforce by approximately 20% or 528 people.

You should never use your pet names for employees in a layoff announcement. It makes an otherwise serious announcement seem tone-deaf.

I also kinda wonder if they accidentally removed a user or some credential that has the permissions needed to make things public again, a TF change could involve both the public/private change and user account changes. Could be a bit to look up an admin account to fix things.

My experience with non-AWS providers in TF is that they're less maintained and buggy - in theory this should be easy, but people seem very afraid of TF and I can picture this getting chaotic.

But you're quite right that if they're comfortable enough, they should go into S3 and get a statefile they were happy with!

I would bet money that Elastic uses a Terraform provider for Github and they marked repos private in an automated way, and the reverse API operation doesn't function in the same way.

It's possible that any delay is them trying to figure out how to get Terraform back to a good state rather than making the repos public being this inherently hard thing.

It’s useful to have a vision like this. Not because everyone will hit all these criteria. (If you do, I’d love to steal your hiring process.)

I find it useful because coaching people to be better is easier with an ideal like this to point to, that is complete. Lots of managers struggle to put into words why they don’t see an employee as they see themselves. If you have a genuine difference of opinion, you can relate back to something like this. Say your employee does a lot of tickets, and sees themselves as a hyper productive engine on a team. You as a manager see them picking up low-impact tickets and not finishing any features the business asked for end-to-end. The employee wants their productivity to be recognized. You want your employee to be more focused on a single business outcome rather than seeing a high number of tickets in the done column.

Some people (especially neurodivergent people) really “get it” when they have a pre-read they can think on and apply to their situation. A blog post like this is nice because you can have the employee read a bullet and then talk about how it applies to their situation in a 1:1.

In the past, I’ve really wanted better strategies for making this work organizationally.

If you say, “we have CRUD apps and some event-driven, truly bespoke stuff,” you have the problem of devs not wanting to work on the CRUD apps and only wanting the bespoke stuff.

Things I have tried

1. Hire seniors who are jaded by overengineering; they are happier to work on CRUD apps and will do things well.

This is my favorite strategy but there are not as many of these people.

2. Have contractors do CRUD and full-time employees work on core features.

This can create an uncomfortable culture split and if you’re not confident in managing contractors, can mean delays or having the customer-facing stuff look sloppy. I’ve also found some contract shops are not set up well to do security-critical things that aren’t a differentiator, like auth.

3. Pay for non-core stuff through vendors, and have employees focus on core things and a smaller set of non-differentiators.

This requires higher vendor spend but is probably the easiest to have a consistent culture and hiring setup.

The European cultures of Spanish/French were equally early (and earlier) to California. The author touches on them but dismisses them a bit.

I get the sense that this analysis is just based on the author’s knowledge of English cultures of the period? I think Spanish/French culture had influence too, it’s just that the author didn’t do enough research on them to be able to include them in their model.

If you want an example of a language where the exact opposite advice is taken at all times (with all the pitfalls described in this blog post), give Clojure a whirl.

Kinda depends on how much you value inductive vs. deductive reasoning, but the authors make the deductive case that:

- There's strong incentives to misreport in these areas (the compelling example from Sardinia was that the person is alive for the purposes of pension fraud, but really dead)

- People who are incentivized to report people being older than they are will do so

And the inductive case relies on data, which is presumed to be totally flawed because of the misaligned incentives.

The security vulnerabilities that affect my PII kinda contradict the idea that the kernel is an implementation detail. It’s not an implementation detail if I do, in fact, have to care about it.

Do you think that we’ve discovered and cataloged the sum total of milder pathogens that protect against worse illnesses in a lab? Got all mine means “got all the ones we currently know enough about to develop a vaccine for.”

As much as I hate the amount of illnesses my kids get from daycare - maybe this shouldn't be a goal? In workplaces, hospitals, etc. we definitely want to reduce illness. In daycares, the kids are all building their immune systems and the contact with germs is a vital part of that.

I definitely understand the teachers don’t want to be sick, and it’s a hardship on parents to keep kids home, so it’s not all about the kids’ health. But the kids’ health might be better served by letting them get more minor illnesses, not less.

No, I am not upset about anything the DNC has done. Trump has more and nastier court cases (including a pedophilia case) than the convicted felony. The RNC is abhorrent and the two cannot be compared.

No, I am not feeling riled up about this. I’m much more riled up about the state of the executive branch from 2016-2020.

For $150/hour, I will spend today consulting for businesses who need someone in the St Louis area to go reboot a remote workers’ machine.

The one piece of help I will offer in FizzBuzz is a reminder of how you can use the modulo operator, for exactly that reason.

This is one of the thing you pay engineering managers for - someone should jump in and ask the senior people to pause while we go to others’ points.

I think this gets at my ick with IAM. If AWS knows it needs “s3:ListBuckets” for a call to function, why doesn’t giving permissions on that call just imply that I gave it the “s3:ListBuckets” permission too?

For enterprises that genuinely want the finer grained control, let them express that they want to opt out of that implicitness in the policy document.

I don't know if I'm bought in on their reasons why this is the case, but the observations seem true to me. The times I've moved the fastest are also the times when everyone is not fully busy.

This is true of large teams, small teams, and every type in between, in my experience.

> The right to freely travel

You would need to change that to "the right to freely travel by any method of my choosing" for your point to stand. If you don't believe in the right to commute by tank, it's pretty easy to see that some communities may not extend a right to freely travel by car.

Airports post-911 have a robust security setup with only a few clearly defined entrances and exits, with a person stationed at each one, in addition to whatever team you have brought with you. If you’re arrested at the terminal, they have likely already swept you for weapons without raising your suspicions already.

I think their tool is designed for working professionals on laptops - seems like analysts doing traditional business insights reporting. Perhaps there's a world where that target audience is using their phones, but I doubt it's a use case they're too concerned with.

I would be worried about this approach for something that doesn't need to expose a SQL interface though! It's not the right fit for most webapps.