hckrnws
Very weird errors in the site, is this AI generated?
When you click on the 2 year old options, it says "6 Activities for Kids Aged 2" despite there being 19 shown, and the text begins "Nine-year-olds are full of ideas..."
The images are good but kinda off... e.g. for https://offline.kids/activity/water-play-tub/ the kid and tub are floating in a even larger body of water, and for https://offline.kids/activity/fabric-sensory-tunnel/ there is a magical rigid blanket-tunnel.
It looks like most of the blog posts for the site were also generated by ChatGPT. In the most recent article [0], all of the provided links include "?utm_source=chatgpt.com"
In "How to Reduce Kids’ Screen Time – Without the Tantrums" [1], half of the expert quotes were misquoted or could not be confirmed.
[0] https://offline.kids/teaching-kids-to-lose-gracefully/ [1] https://offline.kids/how-to-reduce-kids-screen-time-without-...
Good catch. Yeah, definitely AI-generated. The text in the images are in that unmistakeable ChatGPT 4o image generation font that's weirdly kinda fat and oddly kerned.
On AI's struggles with hands: do humans have four or five fingers? Why not both!? https://offline.kids/activity/diy-jigsaw-puzzle/
>On AI's struggles with hands: do humans have four or five fingers? Why not both!? https://offline.kids/activity/diy-jigsaw-puzzle/
You missed the puzzle itself, where it’s actually three fingers. Kind of impressive degree of inconsistency, honestly.
The images are certainly GPT Image 1 generated. Also, the text on the about page is almost certainly AI generated.
I never fully believed the “em-dashes are an AI giveaway” meme — until I saw that front page.
I'm really sad about this, because I'm a word (and punctuation) nerd and use em-dashes in my own writing, because they're correct, dagnabit. Now I'm reconsidering the practice.
This may cause some psychic damage, but I use en dashes instead for this reason... even though yes, it's only conventionally used for numeric ranges.
Keep using them. Don't let AI steal beautiful em-dashes away from you.
LLMs don't put spaces around their em-dashes.
Looks exactly like images generated by M365 copilot if I don't ask for something specific.
The "illustration" on the front page absolutely smacks of the ai-generates aesthetic, and especially the illustrations in the "posts".
I don't understand what the point of this website is. It's disingenuous, shallow, and artificial. If someone wanted to outsource their relationship with their offspring to a text generator, why wouldn't they just go to ChatGPT directly?
I can't imagine there's much overlap between people who want their kids to have less screen-time, but also have no standards for what replaces the screen time.
all of the pictures are AI-generated, probably from GPT-image-1
The images are probably AI, but I don't think the site itself is. It's a WordPress site, and the about page says the author is a web designer and the link in the footer is to a WordPress agency.
Predatory AI-generated site feeding off parents' anti-screen anxiety, no thanks.
We need Klutz to come back https://en.wikipedia.org/wiki/Klutz_Press
I remember having the juggling one! Thanks for reminding me of Klutz. I'm hoping to finally have kids in the near future and, while I don't want to completely shield them from all tech, I do want to ground them in reality with "real" activities. I may order a bunch of their books in the future.
Just put your kids outside. You don't need anything as a kid to start playing. We used sticks and mud and built ourselves houses and towns and had wars and put on plays and did everything under the sun without toys or anything.
Some trees and dirt will take you a long way providing thousands of hours of fun. As kids we found these big black horned beetles and started a beetle gladiator arena that kept us preoccupied for months at a time feeding and training our biggest beetles. Kids are very creative if you let them be.
Exactly this.
We have so many outdoor toys from footballs and outdoor table tennis tables, to outdoor chalk, sand pits and so on and so forth.
Yet most of the time the boys just want water fights and the girls just want to do cartwheels.
Structured play is definitely important. But unstructured play even more so. It’s amazing what kids can find to entertain themselves when they’re left alone.
The fact that 'join a X group/class/workshop' shows up so often does make this a lot less useful. When you're looking for "easy-to-set-up activities for kids of all ages" a sign up process with some external organization isn't really 'it'.
Some context from the dev's blog (https://highrise.digital/blog/building-offline-kids-a-direct...
---
"Within the last few weeks, Mark and I have built and launched Offline.Kids.
It’s a website to help parents reconnect with their kids and for kids to reconnect with the world around them.
Offline.Kids is directory of screen-free activities for all ages. Each activity is categorised so that parents can find appropriate activities for their situation.
For example, you can find:
quick, clean activities for a 6 year olds outdoor kids activities that take 1-2 hours low energy indoor crafts We built the site off the back of our new directory landing page plugin (catchy name still in progress!). It instantly creates thousands of SEO friendly landing pages for the activities. It’s early days, but Google is successfully indexing the pages and we’ll see how the rankings change over time.
So, if you’re looking for screen-free activities for your kids, check out the website, and share with anyone you think might find it useful!"
Interestingly, for it being built as part of the promotional effort for the authors's LLM-based WordPress WWW-site generator that autocreates (in their promotional blurb's own words) thousands of topic pages instantly for search-engine "optimization", this has been submitted onto http://url.town as one of only two entries in that site's "Family" section. This promotion clearly worked.
* https://highrise.digital/blog/building-offline-kids-a-direct...
https://offline.kids/activity/indoor-obstacle-course/
Did anyone even review these AI generated posts before publishing? It's one thing to publish something you didn't write, but it's another thing to publish something you didn't even bother to read:
This activity for ages 3 to 10:
> Instructions
> 1. Clear a safe space in your home
> 2. Set up crawling sections under chairs
> 3. Create jumping stations with pillows
> 4. Make balance beams with rope
> 5. Design tunnels with boxes
First, you should not be leaving children unattended around string or rope (the materials listed here). It's negligent to have that absent from the safety tips, and it's concerning that knowledge is obscure enough that the text generator wouldn't provide a bullet point with that advice.
But also, how many people have a place where they can "Make balance beams with rope"? What low-to-the-floor fixtures do people commonly have with the sheer strength to make a tightrope for children to walk across?
No one needs to hear their screen reader say "yellow squiggle."
I wonder how much of the site was AI-generated. The images definitely are (kids with different numbers of fingers from each other in the same picture lol).
I really want the "Note from the Founder" to be fully AI generated too, image and everything. Our fully unauthentic web of the future has finally arrived!
It's not like before AI we got a lot of genuine founder stories either ;)
I feel while developers continue to spend their time and energy gatekeeping how people build things, AI is going to continue to enable those people to build what they want. You'd be surprised how little people care about how a product was built and just want it to do the things they need it to do.
The MySpace era internet where anyone can create a page is back and I'm here for it
"You'll eat shit and like it" as a marketing strategy? I thought they fixed that in ChatGPT.
Quality, security, and code that doesn't fall apart later matter. I don't want AI slop children books to be a thing. I hear you on AI making it easier for people to build stuff, but calling valid concerns "gatekeeping" is a bit off.
With that said, I really like this site and the approach!
If those things actually mattered they would be rewarded in the "free marketplace of ideas". But they aren't and never have been. The most wealthy companies in the world aren't wealthy because of the quality of their code. That's why literally zero Very Large Organizations prioritize code quality. Marketing matters far more than quality and the budgets demonstrate this. You're just holding it wrong.
The study linked in OP is already a clear counterexample to your point, though. It's clear from all the slop that quality control is low on the priority list of so-called "builders" using AI. They do the first 20% to get a mockup and then decide it's done.
You could just have read the "about" page. Who cares if there's AI involved, this is a dad who made a thing for his own kids, and opened it up for everyone else. So what if they used AI if it does what they hoped it'd do and their kids like it?
Fine for them, but why would I use it instead of using my own AI for my own kids?
It's like reading other people's chat gpt conversations, not very interesting or useful.
> So what if they used AI
It matters because it's a very strong signal of quality.
I'd say it's a strong signal of (lack of) effort, which then correlates with quality
And, crucially, the quality _is_ low. At a bare minimum, the generated safety bulletpoints should be able to anticipate safety concerns with following the generated activities. The people publishing the posts should read them first and check for safety concerns.
The generated posts don't meet this bare minimum. For example, some posts have activities for toddlers involving string or rope, but do not mention the non-obvious strangulation risks. This website should not have been published in this state.
Comment was deleted :(
Because, if you AI generated the activities too, the site is useless. We can ChatGpt generate activities too and get the same result.
is that true? I thought communicating what a full-sight person would see is important. Accessibility isn't just about text. Closed-captions vs Subtitles.
Close your eyes. Imagine you are using your computer or phone, but that for tedious reasons you may only do so via the extremely lossy KVM that is me. Do you need to hear me say "yellow squiggle?"
The point of accessibility isn't an equivalent experience, which is trivially impossible in any case. The point is to make the material as useful, wherein possible. Especially when everything in the UI costs its user the time of its verbal description, "as useful" very often means ensuring the irrelevant is left out.
This looks like it could be really cool. However there are obvious mistakes pointed out by other comments which makes me distrust the content. I'm sure there are still good ideas in there, I just don't think they would be as creative as I would likr
It's funny to use the online to help us with activities our kids can do offline.
Phone obsessed parent looking at website while kids stand with rocks in hand by lake, not knowing what to do. "It says you throw the rocks into the water and have fun."
To think, it's possible this entire thing was vibe coded.
it 100% was. i wonder in 2-3 years if it will start being harder to tell.
It is already pretty hard to tell.
I'm honestly shocked how hard it is to prove that the site is ai generated.
There is a very strong ai-vibe, but to find proof in the pictures is hard on most of them (not the pizza one, that one looks awful).
I'm shocked it didn't occur to me to wonder whether it had been AI-extruded, until I saw everyone else talking about it. The thought of passing on advice, which I myself first received about a generation ago, really was as far as I got.
On the other hand, I have yet to take on a client this year, and so my perspective on AI developments at the IC level is mildly idiosyncratic; I know a grain of salt is required, but not how large a one, and my own experiments reveal a radically different set of capabilities and drawbacks than the technology's manic or epistaxic boosters like to describe.
Judging by what I'm seeing lately here and elsewhere, the workaday tech industry world must be developing into a genuinely horrifying grind, and I'm glad to be out of it.
fyi your (?) css is messed up for the 'Show activities by...' images on a viewport that's the size of a macbook pro 14" display split in half vertically
it should be a printable book.
Comment was deleted :(
Can we please stop using “screens” as a pejorative?
There are plenty of examples where a screen provides a better and more enriching/edifying experience than dead trees, etc
It's true that screens are helpful. But at least in many cases, they are making life worse. It's not that there aren't benefits, but in many specific domains, the negatives outweight the positives. This is true for children because they no longer have mental space to think for themselves, and it is true for many adults as well who spend 8+ hours in front of them. It might make the enterprises they work for more efficient, but it doesn't necessarily make the work enviroment better.
So I think there is sense to use "screens" in the pejorative sense. They are quite irritiating.
The misattribution to screens instead of antisocial behavior is what's making things worse. My children are fine despite having basically unlimited "screen" time. It's not my fault other parent's are incapable of parenting and rely on heavy handed legislation to protect their children from themselves. My children's lives are richer for having those "screens". They are far more curious and explore more than many of their peers do thanks to them. I think the pejorative of "screens" is far better spelled as "bad parents". But bad parents won't ever acknowledge that. Especially bad parents on HN thanks to Dunning-Kruger.
It's quite frankly ridiculous that people who associate with "hackers" want to isolate their children socially and technologically until some magical time that their mind is ready for controversy. As if the generations most capable of "hacking" weren't doing things far beyond their parent's understanding. Half the "hackers" here wouldn't be a fraction as successful if the censorship and restricted access they wanted were present when they were kids. Ladder pullers are some of the worst people in society.
Where do you see "screens" used in the pejorative?
On the linked page:
> Discover simple, screen-free activities
The implication that screens are bad is obvious to normal people.
The evidence is less clear: https://www.bbc.co.uk/news/articles/c9d0l40v551o
> The late Steve Jobs famously didn't let his own children have iPads when they were young
Is this from his biography or something? I haven't read it. But the iPad came out in 2010 and Jobs passed away in 2011. I'm not sure how the timeline works there.
This claim has been passed around for years and comes from an article in 2011.
>“So, your kids must love the iPad?” I asked Mr. Jobs, trying to change the subject. The company’s first tablet was just hitting the shelves. “They haven’t used it,” he told me. “We limit how much technology our kids use at home.”
Thanks! It makes a lot more sense in context.
I don't think pejorative means what you think it means
You made me doubt myself so I looked it up:
> expressing contempt or disapproval.
Exactly what I thought. "Screen-free" is clearly implying disapproval of screen time. What do you think pejorative means?
> "Screen-free" is clearly implying disapproval of screen time.
This is entirely an assumption on your part. Just because parents are looking for screen-free activities doesn't mean they're anti-screen. They're two totally different things. Most parents want to balance screen time with screen-free time. This doesn't imply anything. When you see a gluten-free option on a menu, do you feel so attacked? While some people may be so gluten-free that they impose their preachy anti-bread beliefs on others, most folks don't and are either looking to avoid wheat or have an allergy.
Nice work on offline.kids and kudos for tackling screen-free play. I've been playing in a similar space. If you want something complementary for the inevitable “but why?” moments, you might like StudyTurtle Ask (https://studyturtle.com/ask). It’s a free, no-signup AI Q&A tuned for 3–9 year-olds with:
Strict age calibration (matching phrasing and examples to each developmental level)
Concrete analogies (“volcanoes are like shaken soda bottles”) and kitchen-table experiments you can actually do
This is just an AI generate advertisement in the comments for a link to an AI generated site.
i wrote it...
I believe you but this only shows how hard it is to tell ai generated text apart.
Crafted by Rajat
Source Code