This relates to why a college instructor I know would stop reading submissions if they went beyond the word count limit. The true challenge wasn't showing what you know; rather, it was showing you could carefully select from the near-infinite knowledge space and concisely to respond to the prompt

Pavel@spavel.bsky.social · 5mo ago

This is not a shitpost, I'm serious. Adding things is so easy that any computer can do it. Determining whether it improves the whole (and leaving the whole intact when removing the excess) is purely human, because it requires a point of view.

free open source software needs fewer engineers and more designers and product people

I am the proud architect of a private cloud going into production with not 1 but *3* different storage backends on EOL hardware cobbled together to meet the cloud's storage demands until we can get the real production storage backend in place that was delayed due to shenanigans.

have you ever read one of my Discworld QOTDs and thought, hey, I should really get around to reading those books? good news! the full series is back on Humble Bundle - it’s Kobo but it’s pretty easily to transfer those to your preferred e-reader you won’t regret it

Humble Book Bundle: Best of Humble Bundle: Terry Pratchett's Discworld by HarperCollins Encore 2026

Experience Sir Terry Practchett’s beloved Discworld series! Get 35+ books covering all corners of this comic fantasy universe and help support Room to Read!

humblebundle.com

Union workers solved the trolley problem, you're welcome

olive oil is a legitimate interest

Angeltrix Point Devil 🔜 StratosFur@angeltaz.dev · 6mo ago

@machete.gay

The thing I have come around to is that educated people, and aspirationally this means everyone, should be *literate* in the full range of fields, from technical through humanities. We can’t all know everything, but all fields should be legible to us. We should know how they build and use knowledge.

Faine Greenwood@faineg.bsky.social · 6mo ago

I should also note that I am someone who has humanities degrees and who *also* has ended up really enjoying doing a lot of STEM-Type-Work professionally: I’m a big fan of both. And I refuse to play along with the idea that one is superior to the other.

I think a lot of the hypothetical use case for AI (LLM, especially) type tooling is papering over poor customer experiences, poorly designed user experiences, and poorly made things in general. Using an agent to pull info off a site because the site is too difficult to do that task isn't winning...

I wonder if part of the allure for LLMs, esp. for coding, is that it taps into the procrastinator's work-about-work need, but you get to feel like it's work because you get something work-shaped at the end without actually doing the real work...

Gonna give systemd timers a try instead of cron for a job to copy data from old OpenSearch indices into an archive index. systemd timers look like a neat thing. I wish they would have used a better variable name than %i for the unit name in templates, but that's my preference for more verbose names

One neat small tech thing I have recently learned about is that you can capture all of the scrollback buffer from a tmux pane to a file to review later using the "capture-pane" command. I've had the pleasure recently of slowly and painstakingingly dumping a couple hundred old tmux sessions to file

At this stage of things, I feel like LLMs (and more broadly generative AI in its current form) are very much like a missing stair (en.wikipedia.org/wiki/Missing...). Some think LLMs are genius wonders, but then LLMs facilitate real-world, significant harm on people with less power in social groups

Missing stair - Wikipedia

en.wikipedia.org

you’re right! it feels mostly awful to Go Online™, these days. the internet in a sad state, for reason after reason after reason. but there is a way to fix it; to trade the voidful howling for a quiet & joyful song. here’s how to win the war for the soul of the internet, and build the Web We Want.

A Website To End All Websites | Henry From Online

How to win the war for the soul of the internet, and build the Web We Want.

henry.codes

Accidentally wrote a mini essay on LI: Carrot-eaters and the psychological essentialism that's smuggled in when we divide people into technical and not-technical www.linkedin.com/posts/drcath...

There's a study in developmental psychology affectionately referred to as the "carrot-eaters study," at least that's how I always remembered it when I was doing a PhD - I sat in the ...

There's a study in developmental psychology affectionately referred to as the "carrot-eaters study," at least that's how I always remembered it when I was doing a PhD - I sat in the upstairs of a chil...

linkedin.com

A message for you all to carry through the year

I got another bag of Harney and Sons Scottish Morn tea, and I must say that tea is terrific. It is a very nice, solid black tea, especially if you like drinking strong black tea. I also recommend Harney and Sons Milk Oolong tea. That tea is some of the best Oolong tea I have ever had.

I am wrapping up a fractional tech lead role for a small software engineering team, and I felt like sometimes my main role was to be an accountability sink for the engineers on the team. That is, they knew what needing doing, but they needed me to say "yes, let's do that".

I wonder if we shouldn't look at code in determining whether to use a monolithic or divided architecture and rather we should look at the experiences (maybe structures, like in Conway's law terms) of the teams and organizations managing the code...

I'm not up-to-date on the state of the art, but JSON Web Key Sets are still one of my favorite machine-to-machine authentication methods when I need to auth communication between services I develop. Mostly because it's stateless, and I don't have to worry about revoking/running a DB/managing state

I wonder if demystifying technology work would help foster healthier relationships around technology. Sure, it is magic, but it's not magic in the Gandalf/Merlin/wizard sense. Oftentimes in my experience, it's just pushing buttons until the computer-shaped thing does what you need it do

I let audit logging for a software feature get cut in the name of trimming scope for an MVP, but I'm inclined to believe that if you think audit logging would be a useful thing to have at any point it's probably worthwhile to include in the current MVP. I can't say why, but cutting seems incorrect

Software cultures are absolutely deprived of compassionate moments where we get to be human together and experience being on the same side together. I really believe that humanness changes people's lives and I'm not giving up on its power. I have seen it measurably work, again and again.

Some recent posts about focusing on the velocity of shipping code at the expense of quality and other things gives me big "I Love Lucy working at the chocolate factory as increasing amounts of chocolate spills out of the machine" vibes

two women are clearing a queue in a kitchen

Alt: A gif of the I Love Lucy TV show episode where Lucy and her friend Ethel, two adult women, seated at a conveyor belt in a chocolate factory. Individual chocolate candies move down the conveyor belt, and Lucy and Ethel work on wrapping the individual chocolate candies. The rate at which chocolates move down the conveyor belt gradually increases, and Lucy and Ethel desperately attempt to clear chocolates off the conveyor belt by wrapping them more quickly, eating them, and shoving chocolates into their clothes as the situation spirals out of their control.

media.tenor.com

I think one true sign of software engineering skill is when you can solve a problem without any new code. Even better is when you can solve a problem through only talking to people and building better relationships and mutual understanding

I think one lesson folks should take away from using LLMs to make code is that the code wasn't really the important part to begin. Sure, you usually need code, but code is mostly the expression of someone's domain knowledge applied to fixing a particular problem at hand

Honestly, I'd be willing to say Occam's Razor becomes a fallacy or a trap insofar as apparent simplicity isn't the same thing as real simplicity. Sure, pick the thing that appears simpler to you, but be ready to handle any unknown complexity that you missed in making that choice

When I see folks on BlueSky mention some personal success, I make a point to reply to encourage and congratulate. It doesn't matter to me that they are strangers and I may never hear about their life again. What matters is that for the price of one reply, someone knows they're seen and supported.

As it turns out, having 16000+ filesystem mounts on a single Linux machine is not a good idea. No one did that on purpose. It appears to be a configuration bug in Kolla-Ansible for OpenStack that mounts a Docker volume inside of another Docker volume. Froze up systemctl something fierce

I saw someone on here talking about how metacognition is the way to save yourself from falling for LLM hallucinations. I think Baldur Bjarnason made a compelling case how that can precisely make you *more* vulnerable rather than keep you safe: www.baldurbjarnason.com/2025/trustin...

Trusting your own judgement on ‘AI’ is a huge risk

Writing at the end of the world, from Hveragerði, Iceland

baldurbjarnason.com

Remember, hit www.eff.org/age for a full set of resources about age verification: Why it doesn't (and can't) work, why it hurts everyone's privacy — including kids, and what we can all do to push back.

Age Verification and Age Gating: Resource Hub

Age verification (or age-gating) laws generally require online services to check, estimate, or verify all users’ ages—often through invasive tools like ID checks, biometric scans, or other dubious “ag...

eff.org