sergiomattei

This is my fave article on the subject:

https://evilmartians.com/chronicles/oklch-in-css-why-quit-rg...

Does it matter?

An example from last night. If I Google “how much is 7KwH”, I get a list of calculator websites for converting units.

If I ask an LLM, it tells me “your usage is pretty low for an apartment, good job!”

The latter is exactly the question I had in mind. Not even close.

> But would it have improved my comprehension? The research says no.

I don’t know, I find myself doing things I would’ve never done before AI.

I find myself doing more projects outside of work.

I discover new problem domains and are less afraid of tackling the unknown.

I code in languages I don’t know and start learning how they work.

I can get anything explained to me, at any time, in a somewhat coherent manner, by a tutor that won’t get tired or annoyed.

Look— I’m not exactly thinking AI is free of criticism, but it’s definitely a revolution in access to information and at least to me that’s a big deal.

Can you list some of these reasons?

My impression is they're being cancelled in favor of full internal adoption of Copilot CLI, which has got much better over the past few months.

AM is critical for emergency scenarios. When Hurricane Maria hit Puerto Rico, all our infrastructure was completely devastated.

The only way to receive news or bulletins for weeks was just one remaining AM radio station that kept broadcasting even as the storm hit and their building began to flood.

Why is Ramp even building a sheets product? That's the question zero that popped up to my head.

> Have the courage to go slowly, especially when everyone else is telling you that you need to go fast and cut corners.

I've been struggling to figure out what "slower" would look like when working in industry. If everyone's working 2x faster, how do you slow down meaningfully without getting axed?

This is beautiful product design.

I will say, I do wish there was a conversation list when looking at folders--having conversations listed on the sidebar can get a bit busy.

Zipped up tightly? This has been all over mainstream news outlets, social media, everywhere.

We are commenting on a submission linked to the New York Times.

I built my own harness on Elixir/Erlang[0]. It's very nice, but I see why TypeScript is a popular choice.

No serialization/JSON-RPC layer between a TS CLI and Elixir server. TS TUI libraries utilities are really nice (I rewrote the Elixir-based CLI prototype as it was slowing me down). Easy to extend with custom tools without having to write them in Elixir, which can be intimidating.

But you're right that Erlang's computing vision lends itself super well to this problem space.

[1]: https://github.com/matteing/opal

No shade, I think it looks cool and will likely use it, but next time maybe disclose that you’re the founder?

This looks nice, congrats on the launch. I'm trying this at work on Monday, I suffer from this problem for tons of MCP tools that are MCP only, not CLI, and completely fill my context window.

First thoughts: it seems the broader community is moving towards Agent Skills as a "replacement" for MCPs to tackle the context pollution problem.

Agent harnesses like Pi don't ship with MCP support as an intentional design choice. MCP servers[0] are being rewritten as pure CLIs in order to support this new scenario.

Thoughts on this?

[0]: https://github.com/microsoft/playwright-cli

Papers like these are much needed bucket of ice water. We antropomorphize these systems too much.

Skimming through conclusions and results, the authors conclude that LLMs exhibit failures across many axes we'd find to be demonstrative of AGI. Moral reasoning, simple things like counting that a toddler can do, etc. They're just not human and you can reasonably hypothesize most of these failures stem from their nature as next-token predictors that happen to usually do what you want.

So. If you've got OpenClaw running and thinking you've got Jarvis from Iron Man, this is probably a good read to ground yourself.

Note there's a GitHub repo compiling these failures from the authors: https://github.com/Peiyang-Song/Awesome-LLM-Reasoning-Failur...

My small agent harness[0] does this as well.

The tasks tool is designed to validate a DAG as input, whose non-blocked tasks become cheap parallel subagent spawns using Erlang/OTP.

It works quite well. The only problem I’ve faced is getting it to break down tasks using the tool consistently. I guess it might be a matter of experimenting further with the system prompt.

[1]: https://github.com/matteing/opal

This is probably the most beautiful homepage + docs combo I’ve ever seen. The copy is awesome too. It feels human.

Great work.

I don't know. LLMs are great at writing code; but you have to have the right ideas to get decent output.

I spend tons of time handholding LLMs--they're not a replacement for thinking. If you give them a closed-loop problem where it's easy to experiment and check for correctness, then sure. But many problems are open-loop where there's no clear benchmark.

LLMs are powerful if you have the right ideas. Input = output. Otherwise you get slop that breaks often and barely gets the job done, full of hallucinations and incorrect reasoning. Because they can't think for you.

Seeing how things are moving, I'm expecting for compute requirements to go down over a longer time horizon, as most technologies do.

I'd rather spend my time preparing for this new world now.

I feel the same way.

> That includes code outside of the happy path, like error handling and input validation. But also other typing exercises like processing an entity with 10 different types, where each type must be handled separately. Or propagating one property through the system on 5 different types in multiple layers.

With AI, I feel I'm less caught up in the minutia of programming and have more cognitive space for the fun parts: engineering systems, designing interfaces and improving parts of a codebase.

I don't mind this new world. I was never too attached to my ability to pump out boilerplate at a rapid pace. What I like is engineering and this new AI world allows me to explore new approaches and connect ideas faster than I've ever been able to before.

Dude, this is awesome. El Nuevo Dia on the Wii is peak bori brain. :)

Real courageous from that guy calling someone a "fat uncle" on a Twitter thread. Could've applied that same energy IRL and told him to tone it down.

Seriously, this is as easy as tapping the dude next door and telling him to tone that volume down.

Negative social skills on that Twitter thread

Thanks for the link.

In the interest of saving myself an hour of time uploading a video, I’ll attest that yes—that street view is as “average day on Pike” as it gets.

To be clear, there are homeless who walk around the area… and Capitol Hill isn’t exactly the nicest area these days. 3rd and Pike isn’t nice. But Seattle in 2025 isn’t real-life World War Z.

Parent commenter should visit sometime.

I agree, Chinatown/ID took a turn for the worst unfortunately.

I’m so glad you mention “going up Pike towards Capitol Hill” because I live right on the section where Cap Hill & First Hill starts.

My lived experience is that I could walk the length of Pike from Bellevue Ave towards the waterfront and see two-three tents at most. Homeless yes, but encampments? “Overrun”? “Tent city”?

It’s so easy to make hyperbolic statements and a pain to prove them wrong. I can see how you get away with spouting this nonsense—and people believe it.

Let me know if anyone wants a YouTube video or something, I’ll be glad to take a fun evening stroll and prove you wrong.

Deflecting is a fun game, but I won’t play it. I’m challenging your hyperbole here—on the other hand, you’re making assumptions about me.

Explain to me how “the entire downtown area is just homeless tents”. By all means bring some proof of Pike Place being “overrun with druggies”. Get real.

> The entire downtown area is just homeless camps.

This isn’t even remotely close to true.

> You can't browse around outside Westlake Center without being bombarded and accosted by aggressive panhandlers. Even the iconic Pike Place Market was overrun with druggies.

I live on Pike Street. There’s homeless, but it’s not “overrun” and for the year+half I’ve lived here I haven’t been “accosted by aggressive panhandlers”. These are areas with constant foot traffic.

Yes 3rd & Pike is bad, but it’s improved since then. Late at night it’s not the best—but it’s never been.

If this is truly your experience, I urge you to visit again. Seattle is a large and beautiful city with a lot to offer.

To clarify, this isn't based on user research or survey results, as other Nielsen Group writeups are.

It's a senior editor's opinion on the UI of iOS 26.

And I’m so glad to see them take a big, glorious L after the changes were back-pedaled a day later.