Powershell Users Like To Vomit
In a stunning new study, PowerShell users insist that they like to vomit. How can this be? James Neno says, “PowerShell syntax is too wordy”, a quip about how PowerShell cmdlets can sometimes feel verbose. Similarly, Novel Logan said, “I have a lot of questions for those that chose vomiting lol”, indicating an apparently common sentiment about PowerShell users. Another respondant, going by the name “Fred” said, “[bash] makes me feel powerful and runs on all platforms […]”, a somewhat gross allusion to Bash user’s preference in contrast to PowerShell.
We caught up with Tim Kellogg, CEO of dentropy, he said:
This is certainly a surprising result. I wouldn’t have predicted this. In fact, I’m not sure what I would have predicted at all. But it’s data, we can see the data and it’s crystal clear. You can’t question the data.
You can’t question the data, quite true Tim, quite true. The Data Gods have spoken, and their wisdom sends a clear message: PowerShell users prefer vomitting, Bash users perfer diarrhea.
You can’t question the data.
Cut! Cut! Cut!
Yes, you absolutely can question the data, please question the data.
The Poll
I manufactured the data. Sort of. Given a large enough yet also small enough sample size, a bias will form in the data. It’s not guaranteed, but close enough.
I sent a poll out on 🦋 Bluesky asking people if they preferred Bash or PowerShell, and another question about if they preferred diarrhea or vomiting. I cut it off at 10 respondants, the perfect number to show a clear trend, but not long enough for the Law of Large Numbers to take over and indicate a ridiculous result.
(btw be suspicious of social media polls, they’re not a random sample, they’re a poll of that person’s followers)
The Title
Technically there was a much stronger correlation with bash users, but you gotta admit that people who prefer vomitting are sus. Plus, people who prefer PowerShell are in the minority and there’s this weird tension such that focusing on PowerShell is guaranteed to bring heat to the conversation. The clicks don’t bait themselves.
The Quotes
For good measure, I took some of the quotes very out of context; I made one say the opposite of what it actually said by selectively quoting. Plus I quoted myself, lol.
The Data Gods Demand Obedience
Is data better than pure vibes?
I honestly don’t know. In a lot of ways, data is worse than vibes because it triggers a sort of virtue cycle that makes people stop questioning. Shuts their brains off.
But that’s not a problem with having data, it’s a problem with not questioning.
Carl Bergstrom & Jevin West wrote a book, “Calling Bullshit”, where they discuss several different categories of “bullshit”, or data that appears to be worth paying attention to but is actually lying. They also teach a class by the same title at University of Washington.
Data is good — for starting a conversation. Most mistakes with data are made in how it’s interpreted. Interpreting data is challenging, but like all challenges, it gets easier with practice. You can practice by talking about it with other people.
What should you talk about?
Doubt everything. Doubt yourself and your own biases, doubt other people’s biases, doubt the author’s interpretation of data. Doubt everything.
When questioning other’s interpretations, ask yourself if there’s another plausible interpretation. List them all. Practice makes perfect.
The AI Generation
Here’s the other thing: We can’t trust anything anyway. Get used to it.
Google places AI generated answers (that regularly hallucinate) before actual results. AI generated images and video now effortlessly impersonate elected officials. Politicians… (heh, you know).
Skepticism isn’t just your friend, it’s your only hope.
One day, you’ll be sitting in your living room and your daughter comes over after buying her first home at age 45, as generation alpha does. Talking on the phone she says, “he’s ranting again about AI hallucinations, I’m worried about him, it’s like he never learned to be skeptical.” Because kids learn stuff.
It might seem like a sudden change, but it’s really not. It’s been developing for 20 years. It’s never too late to learn.
Be skeptical. Talk about the data.