Skip to main content


The most underrated skill in tech engineering is the ability, or maybe the *willingness*, to be an idiot. Sometimes the best way to debug a problem is to forget everything you think you know and re-test the most basic things from ground zero. Sometimes the only way to design a good interface is to look at what you've made through eyes that have no idea what they're looking at. You know too much. Your users don't know anything. Be a user. If you can't be a user how can you make a thing users want
in reply to mcc

Often I complain about things in software and get baffled replies like, okay, but you can fix this arcane problem with [arcane solution], you know that right? And possibly I did know it but complain anyway because *I shouldn't have had to know it*. Your software shouldn't have needed a Theory of Operation. Your software shouldn't have required *learning*. Let me reserve my executive function for performing Useful Tasks and not like, configuring interface minutae. For a moment, let me be an idiot
in reply to mcc

The “assumption” strikes again - mastodon.social/@dahukanna/111…
in reply to mcc

an analogy I've used with the software team for the app I work with.

Imagine you have two microwaves in the office lunchroom on the same circuit and using both at the same time trips the breaker. Do you:

1. put a warning in welcome package when they join and yell at them when they forget or missed reading it?

2. put a note on the door of one of the microwaves warning people?

or

3. simply remove one of the microwaves from the lunchroom

Shocking how much developers insist on option 1.

in reply to Bee O'Problem

@beeoproblem @distinct
That would require approval and budget to schedule a visit by a certified and approved electrical contractor. While we're waiting for that. . .
in reply to Dec.tar.gz

@dec23k @beeoproblem @distinct Sometimes a very important part of engineering is to look at a problem and go "I'm going to ignore this, and solve a *different problem*." Sometimes this is a good thing. Sometimes this turns an unsolvable problem into a solvable one! But also sometimes the best way to solve a problem involves changing the problem in a way that involves increasing the budget. And that is the one thing an engineer is almost never allowed to do…
in reply to mcc

@dec23k @beeoproblem @distinct So this can make being an engineer feel very surreal sometimes…

Engineer: "We can solve this problem instantly by spending $10,000"

Manager: "We don't have $10,000 in budget for capital expenditures. Work around it"

Engineer: *Spends two weeks working around it*

Engineer: *Has a total compensation package that costs the company the equivalent of $10,000 for every two weeks of labor*

Capitalism: I am the most efficient way to allocate resources within a society

in reply to mcc

@dec23k @beeoproblem @distinct There's the related but distinct problem that an engineer may be required to spend months working to justify spending one week fixing a problem. Even aside from not being allowed to spend budgets, engineers who understand the technical issues with a project aren't often allowed to have input into how those issues are prioritized.

An efficient allocation of resources, as you say.

in reply to mcc

@dec23k @beeoproblem @distinct Same shit as so many places that are weirdly stingy about buying new hardware.

Engineer with 120k/yr+ compensation package: “Hey I'm spending several hours a week fighting with this gadget, can we buy a $1,000 replacement?”
Manager who can't divide 120,000 by 2080: “No, we don't have any budget for capital outlays”
Other manager: “We're not progressing as fast as we need to for our schedules, better hire another engineer”

in reply to mcc

I keep telling people that writing documentation is often the best way to realise an API/tool you’ve designed makes no sense or could be significantly improved with small changes that you never realise until you describe it. Also best results are when you do this after not touching the project for a month
in reply to The Seven Voyages Of Steve

@sinbad @inthehands This is why it’s such a foolish and tragic mistake to delegate the job to AI, but guess what everyone is going to do anyway. Writing is teaching and teaching is understanding and you can’t solve problems you don’t understand.
in reply to Dan Wineman

@dwineman @sinbad ICYMI, a favorite of mine that ultimately gets at this same sort of line of thought: jenniferplusplus.com/losing-th…