Q&A highlight
Posted20 days ago
Key Takeaways
lmao
I think AI is going to greatly decrease the quality of the internet in the coming years. What we have now, as bad as it is, will be remembered as being very good in 2035. AI will clog this place up.
Eventually there will be a serious attempt to make an internet where each user is a verified human person, to sieve out the AI slop, but I don't think that will succeed and anonymity is part of what makes the internet good.
One solution for some - but not all - of the problems we have is to go back to small communities, where each user knows one another, talking in their own servers, like BBSes often used to be. Discord leads the way here.
* Businesses want to maximize shareholder value
* Those running websites want to do as much SEO-slop as possible to appear first
* Content creators need to maximize views, which means rage-bait, clickbait, etc.
* Addictive content = more time spent, more ads seen
You can't 'fix' the internet. The internet, like many things, is a tool. Shareholders and individual actors are only interested in maximizing their own gains - and so use this tool to that purpose - regardless of any negative effects on the whole. That's how humans operate in general (with rare exceptions).
Everything would have to be self-hosted. No ads would be allowed anywhere. No business would be allowed to build anything, just users. Some kind of super-admin would have to have to power to perma-ban any website or user that breaks the rules.
But you'd still have the problem of people who aren't directly monitizing things, like influencers. You'd have bots. You'd have subtle ads that don't quite appear to be ads, or users writing fake testimonials.
Still, despite its obvious flaws, it would be cool to see someone try to build such a non-commerical internet someday. I wish them luck.
But trying to make a 'non-monetizable' internet is an oxymoron. If it has the ability to allow people to communicate, then it can be used to sell things.
You can't have an exchange of information without that information potentially containing garbage designed to make someone else money. You can only eliminate spam calls if you decide to get rid of the invention of the telephone.
Still, with intense moderation it is sometimes possible. Wikipedia has a vast amount of information passing through it and has stayed pretty free of monitization - although, certainly some companies have written themselves some pretty positive wiki-pages - in general I would say it is a success.
This is the same problem law has, even at the global scale.
You can't moderate when almost every single person in the chain is a bad actor. Individuals will chip away at any structure or organization day by day, year by year, until they are eventually rewarded.
Obviously, this is still possible and many people do maintain blogs and websites for the love of it, the overwhelming amount of internet users are happy using a handful of giant social networks.
Heck, internet dating used to be looked down on as something so pathetic only complete losers did it. The world and the internet has evolved and as much as I miss the wild west days where it felt like anything was possible, the giant corps have taken over and people are too scared to say the word suicide on YouTube lest they be demonetised, so everyone trembles and mutters unalived and everything gets a little bit worse.
I’m rambling, sorry,
It seems the prevailing pattern is that removing barriers to entry means things get worse. In 2025 there are no barriers left and everyone is online all the time. It’s the end stage of the internet we loved.
In fact the barriers have been destroyed so utterly that we are awash in AI generated content now. The dead internet theory is real. Why are we still here?
You'll really be surprised how quickly YAGNI kicks in for like 99% of the internet.
The internet is dead as we knew it, long live the internet.
The content you want is still out there.
The old internet feels like it is increasingly illegal to run. I mean, youtube circa 2013 basically is. You wouldn't get away with just letting people upload disney movies there anymore, or just letting anyone join the adsense program.
The whole pervasive censorship problem on youtube is downstream of various scandals where a big newspaper wrote a hit piece blaming youtube for letting the wrong people say the wrong things or letting the wrong people see them.
So in this cultural and legal climate, I going to try to avoid any business where people talk to each other, since I don't want to get in trouble if I didn't do censorship hard enough, or didn't know every regulation globally.
if you choose to, good luck. I am just frustrated that people observe that social media is bad, and thus pass laws that raise the regulatory burden and thus stamp out competition.
Not to say that the modern alternatives don't fix a lot of problems, particularly with accessibility to end users, but whats to stop us from going back and creating a new branch from that base that tries to solve the problems without the toxicity? And more importantly, engaging only with that network? (the sad probable answer: Because we're collectively lazy).
what you describe emerged from open protocols and standards being intermediated by middlemen because “regular” people don’t want to learn how to run their own fiber.
All the “regular people” took over what was basically a communications experiment by scientists because attention and as a result commerce moved to that experimental space.
I lived through that as a early net user in 1992 and watched any possibility of the dream get eaten.
The crowding out of the “original intent” is an artifact of transactionalism being the driving factor behind the majority of human action.
Anything that would replace it would eventually succumb to the same fate because the constituent parts are still the same: humans who desire abstracted transactional convenience over coherence and sustainability.
One approach is to simply disappear into a parallel universe designed to somewhat resemble the old internet, but also guard against what it turned into.
The Gemini Protocol was a remarkable attempt at this, and while it isn't (and will probably never be) mainstream, it seems to me to be a little pocket of sanity (although its jettisoning of semantic HTML-like structures pains me greatly).
To forestall any unnecessary commentary on the design and choice of the GemText markup language: Markdown is difficult to parse and not standardised, and a subset of HTML is also not trivial to parse, and still potentially leads to the same problems as exists today with HTML. GemText was well-designed for its goals.
This was not the case for the Internet or the World Wide Web;
Almost all the traffic on the internet goes to one of like, five companies. They're the ones psychologically experimenting on your kids and intentionally making their services addictive and mental illness-inducing. The other 99% of the internet is still mostly fine. If you simply blocked access to the top 1% most visited websites, I suspect that would solve most of the issues people see with "the internet".
Build wholesome small-scale services just for you and the people you care about. I run a little private photo-sharing/messaging service for me and my friends, along with other nice things like a Jellyfin server for media. There's just no incentive to introduce anything bad like you inevitably get with massive for-profit companies. And, these corners of the web are never realistically going to be subject to all the recent regulation like the Online Safety Act, if it's just a small invite-only space.
We shouldn't have tens of millions of people on one platform arguing over what sort of content should and shouldn't be allowed, ruled by big tech oligarchs. Build an area of your own that you can share with your friends!
Unfortunately I think the mobile app store model significantly increases the barriers to random people making their own things. You can teach a young teenager to make a little website that they can stick on GitHub Pages and show off to their friends in an afternoon. A mobile app? Not so much. Especially if it requires possibly buying a Mac to run XCode, and then an Apple Developer account (maybe not that expensive if you're in the tech industry, but if you're a kid who just wants to try it out and show stuff to your friends, it's a pretty big barrier + there's the id verification stuff). It's unsurprising that such a small amount of the software for iOS is free/open-source, compared to platforms where the costs to develop are lower.
Once sites get too large and the mainstream enters, they seem to go downhill quickly. I think the old internet seemed better, because this didn’t really happen. There were pockets of like-minded people, not general platforms trying to be everything to everyone.
In addition to this, get the stuff off the phone. Having these things in your pocket and available all the time doesn’t seem to be good for people. Make the sites painful to use on mobile, to select for users who are willing to engage only with a proper keyboard and mouse, which will drastically reduce low-effort posters, trolls, and excessive engagement.
Of course these things are counter to everyone’s dream of getting rich on the internet. We have to choose. Do we want a healthy, collaborative, and functional internet… or do we want to make a lot of money. It doesn’t seem we can have both.
I suspect you're talking about harm from "social" media and business practices that compromise user experience and wellbeing.
no, seriously, that's the main thing. nothing about "internet" says you have to let ad-motivated recommendation engines fill your gullet.
let's call it "mindful internet": you deliberately choose what to read, even if that means, gasp, seeking out a few quality sources, checking in with a few worthwhile people. "influencers" are so ick precisely because they are an gross side-effect of doomscrolling.
Not affiliated with Hacker News or Y Combinator. We simply enrich the public API with analytics.