No shame in doing something that already exists for yourself. But at this point I think it wouldn't be asked too much to just ask your favorite agent to check first if someone already implemented your idea in the past and then consider if your vision is really distinct enough to warrant anther instance of whatever you want to build.
Maybe that can happen through custom instructions. But I guess, by default the agent will happily do extra work and spend lots of tokens on “redundant” projects.
We're going to enter an era of fragmentation and silos like never imagined before. It'll be terribly wonderful, with everyone getting exactly what they want.
> Ah, Manufacturing Consent—the font that doesn’t just scream “serious journalism,” it calmly editorializes it in 144-point Fraktur while subtly nudging your worldview.
Unfortunately feels too much like an LLM output for me.
Isn't what people pay for with the NYT or other publications, the journalism and editing? Like, the UI isn't why people pay for it.
Given that this seems to just be a re-skin, why a full site that updates hourly, instead of just a browser extension that rewrites the content in-place live?
Fun. I have a visceral dislike for the NYT, so the association is actually a negative, haha — but I could get over it.
The tradeoff of the Hacker Times design is increased information for each story but lower density of stories compared to the HN homepage. The high-density design of the HN page is probably optimal for my habits, even though some of the user-submitted headlines are opaque enough that I miss some things I would otherwise click on.
This is great! Very readable and feels like a nice way to browse articles. I tend to gloss over stuff in the HN wall of text sometimes so this breaks the links up in a useful way. Nice job.
Yeah, it does make the stories a lot more readable, HN just has bullet points of links making it slightly hard to follow everything the only thing the site is missing is a games tab like the NY Times.
This reminds me of an idea I had a while ago; it would be nice if I could receive a paper daily/weekly of interesting articles, updates in the world, think pieces, etc. that weren't as broad as daily news or as opinionated. Just things that people found interesting or things happening around us without the political bias (on either side).
The paper side would just be to anti-digitize my life more, feels like it would be almost like a paper RSS feed.
It wasn't physical, but there was a slick app for the 1st gen of iPad called Early Edition which was basically just curated RSS feeds presented in the style of a newspaper. I was only able to find one image of it. I don't think the app exists anymore, unfortunately.
Kagi has a news site/app that updates once per day with the highlights from various categories of news.
I would imagine one could take something like that, and have program fetch it each morning, format it nicely, and print out a front page overview of what's going on. Just pick up the paper from the printer each morning.
This idea might be fine for a front page, but would fall apart when you start getting into long articles and think pieces. You'd be burning through a lot of paper.
Early versions of Instapaper had a nice way to push out a Kindle "magazine" with recently-captured articles that I loved. This was circa...2010, 2011? An age long past.
If you can track down some decent sources that you like (maybe something like Kagi News, https://www.positive.news/, a good local paper if you have one), you could with little effort vibe up a daily printout and have it waiting for you when you get up each day. For a fun version, there's plenty of cheap thermal printers on aliexpress and similar that'd give you a non-US Letter/A4 version so it doesn't feel like Yet Another Document.
Not exactly what you’re asking for, but a big fan of Hacker Newsletter, which emails you top HN articles. What you’re asking for is a little utopian, HN is the closest thing I’ve found
Back when people still understood that they don't need to know about every single "breaking" news story on the planet the instant it happens.
For a couple of months, I took a digital news break. I only read one national paper on Sundays, and the local paper twice a week. Amazingly, I didn't burst into flames or get stranger dangered. I felt both calm and informed. I plan to resume that pattern after the election.
FOMO is the drug. The tech industry is the dealer.
At least on mobile, the format is much more reminiscent of slashdot.
Man, what a tragic waste of a great .com era community. The whole community self-moderation thing worked ok before the more worthwhile people left. Now they can’t even be arsed to pop by once an hour to remove the multi-page swastika ascii art comments.
It had too high of an influence to integrity ratio, and got infiltrated by motivated actors. That is not an uncommon dynamic. Every democratic forum kind of medium with significance becomes a target for career propagandists and con artists and so on.
This is so cool! Definitely liked the title font :)
I do wonder if this kind of republishing is compliant with the original site? As in, if one can store the content vs. not and just lazily convert live - have been curious about the copyright rules & practices in this area.
Just for your information, if you care about the reachability of your project: this site, like all ChatGPT sites, is blocked in countries that are banned by ChatGPT (e.g., Hong Kong). And because it uses a custom domain, it cannot be automatically proxied as I have set up for ChatGPT domains, which is slightly annoying.
Guess I missed ChatGPT sites, I presume it's exactly what I envision, you can make an accessible website with ChatGPT? If so, how do you know this is one?
washedup is right, of course, that the community is many people with many views (though the point should be made less snarkily). Of course HN disagrees with itself; it's not a self to begin with.
I think there's another factor: it takes time to find the best ways of using such new tools. The community is working this out slowly (as is society generally) and it is naturally a lagging phenomenon.
Seeing generated sites like OP and karpathy's HackerTube (https://news.ycombinator.com/item?id=50319510) which look good and don't have the obvious-LLM sameiness that so many others do, is an interesting step along this way. It's also refreshing. There's no intrinsic reason why generated software should look bad or work poorly.
My point is that people will complain about GenAI but they actually love/enjoy AI generated content or projects as long as they fit what they're looking for.
I also regularly complain about AI generated content. I do that mostly, because it's often either
- extremely uninspired design and UX wise (e.g. the classical beige/serif AI slop website),
- full of obvious and stupid mistakes (e.g. those shitty infographics on social media),
- just mimics the real thing, but misses the actual purpose (e.g. again those shitty infographics on social media)
- or is ridiculously overloaded with features and documentation (e.g. many AI generated software tools on GitHub)
That's repelling to me, because it screams that the human behind it is himself lazy and uninspired. It's bland, overloaded or full of mistakes, because raw output is a commodity.
It feels like a waste of time to even consume something, that the author or developer wasn't willing to put actual effort and thought into.
But that's an artefact of how the author is using the tool and presenting the results.
It's not an inherent property of AI use itself. HackerTimes for example doesn't feel uninspired or bland. And i like it.
A couple of months ago (so probably vibe-coded too):
https://thefrontpage.dev/
https://times.hntrends.net/
https://hackernewstimes.com/
This one even from 6 years ago:
https://www.wolfgangfaust.com/project/paper-hn/ (https://news.ycombinator.com/item?id=23669650, 627 points, 133 comments)
https://hn.algolia.com/?q=hackernews+newspaper
A shame that it broke a few weeks ago. These projects tends to get abandoned after a while.
I like the description for Manufacturing Consent font:
https://fonts.google.com/specimen/Manufacturing+Consent/abou...
https://github.com/google/fonts/blob/main/ofl/manufacturingc...
> Ah, Manufacturing Consent—the font that doesn’t just scream “serious journalism,” it calmly editorializes it in 144-point Fraktur while subtly nudging your worldview.
Unfortunately feels too much like an LLM output for me.
(Which is probably post-ironic in this context?)
Given that this seems to just be a re-skin, why a full site that updates hourly, instead of just a browser extension that rewrites the content in-place live?
The tradeoff of the Hacker Times design is increased information for each story but lower density of stories compared to the HN homepage. The high-density design of the HN page is probably optimal for my habits, even though some of the user-submitted headlines are opaque enough that I miss some things I would otherwise click on.
http://n-gate.com/
The paper side would just be to anti-digitize my life more, feels like it would be almost like a paper RSS feed.
https://imgur.com/a/5sw6zOP
https://web.archive.org/web/20131020160134/https://itunes.ap...
I would imagine one could take something like that, and have program fetch it each morning, format it nicely, and print out a front page overview of what's going on. Just pick up the paper from the printer each morning.
This idea might be fine for a front page, but would fall apart when you start getting into long articles and think pieces. You'd be burning through a lot of paper.
If you can track down some decent sources that you like (maybe something like Kagi News, https://www.positive.news/, a good local paper if you have one), you could with little effort vibe up a daily printout and have it waiting for you when you get up each day. For a fun version, there's plenty of cheap thermal printers on aliexpress and similar that'd give you a non-US Letter/A4 version so it doesn't feel like Yet Another Document.
have you read it recently? Is it still the authentic articles? (as opposed to ads for whatever you were browsing a few minutes earlier.)
Back when people still understood that they don't need to know about every single "breaking" news story on the planet the instant it happens.
For a couple of months, I took a digital news break. I only read one national paper on Sundays, and the local paper twice a week. Amazingly, I didn't burst into flames or get stranger dangered. I felt both calm and informed. I plan to resume that pattern after the election.
FOMO is the drug. The tech industry is the dealer.
Man, what a tragic waste of a great .com era community. The whole community self-moderation thing worked ok before the more worthwhile people left. Now they can’t even be arsed to pop by once an hour to remove the multi-page swastika ascii art comments.
We built a combined books/video version as well in our app library.
https://medium.com/@vektormemory/vektor-2-0-4-hacker-news-bo...
I do wonder if this kind of republishing is compliant with the original site? As in, if one can store the content vs. not and just lazily convert live - have been curious about the copyright rules & practices in this area.
Source: https://youtu.be/O7AKquUtTvs?is=MJjR5eF9UY_JAo12
Georgia Bold is not fit for the purpose, and there's no earthly reason to use it at 55px.
There are so many good serif fonts to choose from on Google Fonts. An embarrassment of riches.
But honestly, the headlines there are refreshing
https://www.nytimes.com/rss
People complain about AI generated content, then proceeds to give a dancing cat AI reel 1 million likes.
People here complain about vibe coded apps, then proceed to upvote this to the front page.
I think there's another factor: it takes time to find the best ways of using such new tools. The community is working this out slowly (as is society generally) and it is naturally a lagging phenomenon.
Seeing generated sites like OP and karpathy's HackerTube (https://news.ycombinator.com/item?id=50319510) which look good and don't have the obvious-LLM sameiness that so many others do, is an interesting step along this way. It's also refreshing. There's no intrinsic reason why generated software should look bad or work poorly.
I also regularly complain about AI generated content. I do that mostly, because it's often either
- extremely uninspired design and UX wise (e.g. the classical beige/serif AI slop website),
- full of obvious and stupid mistakes (e.g. those shitty infographics on social media),
- just mimics the real thing, but misses the actual purpose (e.g. again those shitty infographics on social media)
- or is ridiculously overloaded with features and documentation (e.g. many AI generated software tools on GitHub)
That's repelling to me, because it screams that the human behind it is himself lazy and uninspired. It's bland, overloaded or full of mistakes, because raw output is a commodity.
It feels like a waste of time to even consume something, that the author or developer wasn't willing to put actual effort and thought into.
But that's an artefact of how the author is using the tool and presenting the results.
It's not an inherent property of AI use itself. HackerTimes for example doesn't feel uninspired or bland. And i like it.