News for hackers. Stuff that matters.

PebblesRox

E.g. your FAQ says it doesn’t use cameras but does not say what it does use. And when I got to the footer and clicked the “how it works” link, it seemingly took me back to the top of the page I had just read.
Very cool! I was curious to hear the details of how it worked but couldn’t find them on your landing page. I’m glad to see them here and you might want to include them more prominently on the page itself. I didn’t even realize it was a pair of glasses til I read the above.
You might only be hearing about it now but the warnings have been coming (including by those who are now leading the top labs) for a long time, since before any of those “incentives” were in place.
Also how many ads are we exposed to where we aren’t paying conscious attention to it?

Surely those ads could sometimes cause someone to think or talk about a concept that they might not have otherwise.

If the ad then pops up again and they notice it this time, it will feel really spooky!

I think it’s similar to the phenomenon where you learn a new word and then immediately see it out in the wild somewhere. It grabs your attention because you were just thinking about it.

In the original example of ordering meat from the butcher, I also wonder if it’s possible that the data from that purchase was tracked and fed into the ad system. That seems more plausible to me than the “phone listening” theory.

It was a really strange feeling to read through Asimov and get to Robots and Empire, where the story, having ranged throughout a galactic empire in the far-flung future, follows the characters not just to Earth but to Three Mile Island — a place I had seen before! A place less than an hour a way from my hometown! It made the story feel more connected to my own life in a way that I was not expecting.
Re:AI Tutoring with Khanmigo in a Two-Year School Experiment on Tuesday October 06, 2026 @12:41AM (#49974268)
That sounds like meaning and motivation to me.
Pangram is flagging the full text of the article as AI-assisted.
Re:GLM 5.2 is nearly as accurate as a human book keeper on Thursday July 16, 2026 @05:03PM (#48940205)
Wife here. Pre-agentic coding I had automated a bunch of low-hanging fruit via low code tools (mainly Pipedream and Airtable) but I never had enough time to do everything that I wanted to.

Lately I've really been getting into agentic coding (first with Antigravity and now with Claude Code) and I'm really excited about what I'm now able to accomplish on a very small budget of both money and time! It's been a big boon to my ability to automate things that really ought to be the computer's job.

Plus with LLMs, the scope of what is properly the computer's job has increased by a lot! I've created a couple of Pipedream workflows that send freight bills and customer POs to an LLM to extract data. Now I just have to review and occasionally make a few tweaks vs. having to enter everything by hand.

Since building those, my main focus has been on building better tools for myself and the team, such as a little web-app for sales reps to create and send their own purchase orders vs. having to go through accounting (aka me). The nice thing about vibe coding vs. hooking together a bunch of low-code tools is that I can customize the UI to make it exactly what we need.

My current project is building a back-office app to help streamline the process of managing incoming shipping paperwork. This is a much fuzzier problem than regular invoice processing because it's a bunch of different types of documents (BOLs, packing lists, scale tickets) and a lot of judgment calls to make. I don't expect to hand it all off to the LLM but I plan to have the LLM do the first pass of transcribing the data. Then I'll make the human-review portion as ergonomic as possible for myself.

I have a first draft up-and-running but the default interface that Claude Code came up with is not very good. I'm currently discussing the UI design with Claude in chat mode, narrating my manual process as I do it to help flesh out all the details of all the different sources of info and what decisions need to be made.[0] Once we're done I'll ask for a spec that I can take back to Claude Code for implementation.

[0](https://x.com/CBancos/status/2077793484115755282)

Re:Where to Find the Colors Your Screen Can't Show You on Saturday June 20, 2026 @12:43PM (#48610638)
I think the lack of depth perception in a 2D photo is another big factor in why the magic of a forest can’t be captured. There are so many different layers that all get flattened!
Re:Calvin and Hobbes and the Price of Integrity on Wednesday June 17, 2026 @12:04PM (#48572405)
Thank you for the link to 1/0 — it had a link to Absurd Notions that I had been searching for to read again but couldn’t remember the name of!
Re:U.S. kids no longer reading for pleasure on Friday June 12, 2026 @08:33AM (#48503324)
Oh and the best author from that discovery was hands down Garth Nix. I love the Abhorsen series!
Re:U.S. kids no longer reading for pleasure on Friday June 12, 2026 @08:02AM (#48503035)
In high school I was at a friend’s house. Half of the bookshelf was filled with books I loved and the other half with books I’d never heard of! I felt like a prospector striking gold :) Cool to think that we have something similar on demand now!
Re:U.S. kids no longer reading for pleasure on Friday June 12, 2026 @07:58AM (#48502997)
Kings of the Wyld is a lot of fun. Plus it’s always refreshing to have a middle-aged protagonist instead of a teen or young adult.
Re:Building an HTML-first site doubled our users overnight on Wednesday June 10, 2026 @02:18PM (#48480444)
I don’t like multi-page forms when I’m not able to see at a glance all the info needed to fill out the form. Though I guess if my progress is durably saved that makes it not so bad. The worst is when I fill out a bunch of stuff and then realize it’s asking for info I don’t have, and after I get the info I have to fill everything out again from scratch!
Re:Childhood Computing on Monday May 25, 2026 @11:37AM (#48268105)
We set up a “kid laptop” for our kids (ages 3, 6 and 9) that has a short list of allowed websites and a curated set of installed programs.

We treat it like any other toy: they can pretty much play with it whenever they want for as long as they want. Of course they have to share it between the three of them, so there’s a natural limit there.

Every so often we’ll add something new; most recently I installed SimAnt after we were watching ants in our backyard.

So far we’ve been very happy with this approach!

My impression was she just brings the typewriters into class as a one-day novelty thing per course, not that it becomes the norm for the whole semester. The goal is to give the students a taste of what the old-fashioned way is like, to get them thinking about it.
Re:Desk for people who work at home with a cat on Saturday March 28, 2026 @12:48AM (#47551733)
I tried this for my toddler but unfortunately it fooled me more often than it fooled him.
Re:Desk for people who work at home with a cat on Saturday March 28, 2026 @12:46AM (#47551718)
An example of a pawprint manuscript:

https://art.thewalters.org/object/W.305/

Re:LLM Inevitabilism on Tuesday July 15, 2025 @02:19PM (#44574260)
If you believe this, people believe everything they read by default and have to apply a critical thinking filter on top of it to not believe the thing.

I know I don't have as much of a filter as I ought to!

https://www.lesswrong.com/s/pmHZDpak4NeRLLLCw/p/TiDGXt3WrQwt...

German
Re:Darebee – RPG Fitness on Tuesday July 01, 2025 @09:44AM (#44433874)
I am starting out on one of the non-RPG beginner-friendly programs but I'm looking forward to giving Spellbound a try!
Re:A simple 16x16 dot animation from simple math rules on Monday May 12, 2025 @03:37PM (#43966744)
My 5yo did great on these until he hit #7 and got stuck! I’m helping him with syntax but trying to hold back from giving conceptual hints for now.

I love the instant visual feedback!

I had a really good time with my 8yo just now, reading a story about him and his baby brother.
Another feature I would love would be the ability to get a printable version so I can hand it to my kids. With that it would be great to have an option for a line-by-line translation (maybe side by side so they could cover half the page to hide it) and then another option without a full translation but with a little dictionary at the bottom so they can look up unknown words.

And my preference in general would be for a more literal, word-for-word translation so I can learn what each individual word means.

I like it! I came back today after making a story yesterday and it had not saved my language.

Also, I’m using it in the phone and don’t see a way to get the translation of just one word since I can’t hover.

I find myself doing some of my most creative thinking in the middle of the night when I wake up and can’t get back to sleep.
Re:When Flat Rate Movers Won't Answer Your Calls on Sunday May 04, 2025 @11:52AM (#43887431)
The optimal amount of regret from decluttering is non-zero!

I’ve learned that yes, maybe 1 out of every 100 items I get rid of will turn out to be something I need again in the future. That’s a worthwhile price to pay for the benefit of not having the other 99 items in my life!

Donating to thrift stores is very convenient. And I learned recently that if I have stuff that’s not really nice enough for the thrift store to sell, I can just list it on Facebook Marketplace for free and people will come take it away from my porch! Makes me feel better about getting rid of stuff that still has some use in it, because I’m not just throwing it in the trash.

My mother-in-law and her friends use a system I call clutter laundering. Anything with too much sentimental value to give to a stranger, they pass along to each other. Presumably once the emotional distance is long enough, somebody can actually get rid of it for good! (That’s what I’ve been helping my MIL do with stuff that comes to her at least!)

"There’s status in a neatly tended space that conveys a message of respectable conformity and leisure. There’s also a strong cultural rejection of useful productive things because they’re useful and productive.

"My favorite example of this concept is the humble clothes line. Is it legal to dry your laundry in the sun where you live? In many parts of the country this is expressly forbidden by law and/or private binding agreement. This sort of activity is associated with rural peasants, impoverished slum dwellers, dirty hippies, white trash (or worse), and is at odds with the look and feel of a prosperous community. It might be a scorching day in August but everyone is compelled to operate a mechanical dryer in the house and crank up the air conditioning because anything else is shameful and verboten."

https://www.strongtowns.org/journal/2020/5/6/the-other-pitch...

Older »