$10-20mil in the year 2571 is like $0.15 today

Feels more like a home manager thing to me

You’re supposed to disable source maps in prod?

Asking for a friend

I use a tiny translucent toy turtle. It’s cute, and I can still see the LED of the power button.

Much safer from cats

But why the Random Capitalization?

AI designed it

Probably a flea, based on this: https://brightside.me/articles/10-bug-bites-anyone-should-be-able-to-identify-373860/

Or fly? https://blog.entomologist.net/wp-content/images/?q=what+are+the+different+types+of+bug+bite+marks

Yeah, just comments… with chapters…

Honestly, the developer experience was shit.

They tried to leverage their decades of prior investment and use it as an advantage, but what it actually felt like was a wobbly Jenga tower where every little thing had a caveat and no clear happy path.

Contrast that with iOS, where it felt like they basically started from scratch.

I think Microsoft thought they were lowering the barrier to entry by allowing existing WinForms, ASP.NET, and Silverlight (lol) devs to reuse their stuff, but in practice it made it harder to get started. Every app felt like a legacy codebase from the jump.

Hard to say, actually.

• .NET took an unexpected turn towards cross-platform FOSS
• A third major player in the smartphone market may have abated the enshittificatory forces for a bit longer
• Having a platform that’s consumer-oriented, in contrast to their mostly business-oriented offerings today, might have clued them in to consumer sentiment a little better
• Having a viable path towards profitability outside of enterprise services might have made the all-in gamble on OpenAI less appealing
• Butterfly effect etc.

It’s the problem, but also the strength. That fragmentation allows room to experiment.

It also puts pressure on the underlying protocols/specs to be air-tight. If you have just one implementation to support, you can do whatever. If you have to support 15, all with different goals and constraints, you gotta be pretty damn careful.

So in the end, we get foundational systems that are able to evolve over time instead of needing a breaking-change, ground-up rewrite every 2 years.

Kingslayer85 is an okay name I guess

Tim Curry

12oz mouse

“I always wanted to play in a band”

“Oh yeah? What do you play?”

“I play, uh… with minds. Now go home. But don’t look under your bed”

URL shorteners, AMP? Micro USB?

Edit:

Thinking of things that weren’t made obsolete but just unprofitable…

Physical menus at restaurants, useful search results, human support staff, non-subscription software, open APIs, useful product reviews

My stream of consciousness: “What? Reed isn’t pronounced like led. Oh there’s more here… Ohhh, red is pronounced like leed. Er, reed is pronounced like… uhhh… anyway, I get it.”

Okay but like… the Luddites were right though.

They weren’t opposed to technology. In many cases, they were the ones who built the machines they would later destroy.

They were opposed to letting capital owners dictate how the technology was used. They worried that they would end up working longer hours, in worse conditions, for less pay.

They died (and killed) to prevent this — to the point where destroying a knitting frame was declared a capital offense.

While they did get disbanded eventually, they also laid the groundwork for modern labor rights.

Which is why it’s super disappointing that their name has become a derogatory term for being stuck in the past, when they were ultimately calling for a progressive technological revolution that we have still failed to achieve today.

So here’s the thing… In between the land of “shitty service jobs” and the land of “fully automated luxury” lies the vast desert of “reverse-centaurs”.

Right now, when “AI” takes over 60% of a job, that remaining 40% becomes a brutal dehumanizing gauntlet: the “human-in-the-loop” becomes a peripheral for the computer, manipulated into working at the speed that the computer prefers, like Lucy in the chocolate factory, until they’re used up and replaced. Think Amazon warehouse pickers or drivers.

Part of the problem is that this exploitation is hidden from consumers. When we see a fellow laborer suffering horrible conditions in a public-facing service job, we’re much more likely to throw a fit than when they’re hidden behind a sleek UI.

With no guarantee that we’ll ever make it through to the other side of the desert, I’d be perfectly content to stay on this side of it.

When self checkout started, it was too dumb. It would panic if you breathed on the scale wrong, frequently double-scan items or just have weird bugs.

Then for a minute, it was perfect. They smoothed out the UX, and everything Just Worked™.

Now self checkout is too smart. The camera sees me grab multiple items to scan back-to-back, or sees my kid playing with the bag carousel, and it sets off a shoplifting alarm that the employee has to come over and clear 2-3 times per trip.

So I’ve caught myself adjusting my behavior, like the Amazon drivers that get penalized for singing while they drive because the face-tracking throws an alarm.

If it were just me, I probably wouldn’t think much of it. But then I wonder: Is my daughter going to have to adjust her hands, her posture, her facial expressions… to be acceptable to an ever-present AI observer, for the rest of her life?

That seems to be where we’re headed.

What happens to the misbehavers?