I know Scott Adams is not someone to be promoted for his personal views, but Dilbert is so much a reflection of reality it's unreal.

Same when they interview coaches after a game.

“What do you think went wrong, and what can you do to improve it”

“Well, I think mainly our problem was, we didn’t score more points than the opposing team. That was a major contributing factor to our loss. I think in the future, we need to focus more on scoring more points, and not allowing the opposing team to score more points than us.”

“because of inflation, costs are rising”

The more common way these guys increase profit is by lowering costs, and the only way they've figure out how to do that is by laying off a bunch of people, so this is actually a step in the right direction at least.

Now I'm wondering whether their solution is to appeal to more customers or to raise prices. What am I saying. Of course it's the latter.

Engineers: Hmm, maybe we should get someone with a bit of market knowledge to the table.

MBA: Shit, I have no clue what they're talking about. I need someone who speaks my language.

MBA 2: Man, these engineers really have no clue what we're talking about, huh.

Engineers: removed

This is why Boeing builds planes whose wings regularly fall off.

As in, it's literally the reason. They replaced all of their engineer managers with business managers who don't understand the process.

Jack Welch hollowed out and destroyed one of the largest conglomerates in the world, but he was able to hide the damage until after he retired and he made a ton of money doing it.

So obviously his methods became the standard for the business world, even after GE's collapse and the countless postmortems laying the blame on him.

Y'know what, let's circle back on this, cuz at the end of the day, we're a family here.

Now, give me 50 million dollars.

Also engineer here. Please, listen to engineering. I'm so tired of product coming in with ideas fully detached from reality.

At one job, they got it into their head that "our system has no concept of an account. There's just projects floating around, and nothing unifying them. We need to do a bunch of work to create this". I said to myself, that's crazy. There is an account. Every project has a foreign key relationship with it. It's just not named "account" for some reason.

Listening to me took what could've been a clusterfuck of wasted weeks into a one day find-and-replace project. Personally, I would've just left it with the slightly weird name and called it a win, but I think product needed to feel like they were adding some value somehow.

Or the time they wanted to fully rewrite the internal tool for scheduling work. We had operations people that managed the field workers schedule, using some home-grown tool written years ago and never really updated. They wanted a full rewrite. I talked to the people who actually use the thing and asked them what their biggest pain points were. Looked at the code. Yeah, one of those can be fixed today, the other in a couple days. This doesn't need to be a two month project. We did it my way and operations was delighted.

One time I wasn't in the room, and product and one less good engineer got it into their head that there's no way to tell which work orders go with which set of outputs. They thought that the output just appeared, and you couldn't tell where it came from. Unfortunately, this spun up into a "we need to rewrite the entire system!" project. Some months later (of delivering no value to anyone) there were layoffs, and at great personal cost I was able to convince them that yes, there is a foreign key, and we can make significantly smaller changes to solve the actual problems. I regret not killing that initiative earlier, but I think people wanted it as a big line item on their resume.

That's all startup land.

At the megacorp I worked at, trying to convince management that we should have automated tests is like trying to speak french to someone who only speaks italian. I think they understand some of what I'm saying, maybe, but most of it's not getting through. A good chunk of the IC engineers know the system is bad and has a bunch of "we could improve this in a day" tasks we could do, but management doesn't understand. So we keep having multi-day deploys with "omg it's broken again".

This reads like a loading screen tip

“How good you are at something will dictate how good you are at that thing.”

I know someone who told me (in French) M.B.A stands for "Moins Bon qu'Avant", that translates to "less good than before".

Dont take Flatulia if you're allergic to the ingredients in Flatulia. Ask your doctor about taking Flatulia

Let's parking lot that for now and circle back the next stand-up. Don't forget to Slack the SME taking point on this vision.

Most nonfiction books I have read would have been better written if they were ¼ the amount of words

It's by design.

If you "best guess" up a success based on no actionable guidance, they take all the credit, you only did what you were told, like any common peon, they had the "real vision".

If your best guess falls, well your execution failed their brilliant insight. Not their fault, circumstances failed to get them the right people.

With the experiment that I believe Anthropic made I wouldn't even be sure of that