Neil Gaiman on the future of the internet | Hacker News Reader

Neil Gaiman on the future of the internet

calnewport.com

64 pointsvitabenes41 comments

Worth noting that Neil Gaiman is quite possibly the most active human user of Bluesky on the entire platform. He has absolutely stupendous output, and it's all kind and decent.

Unfortunate.

Why?

Bluesky

Why is Bluesky unfortunate?

Sometimes I wonder if the cozy feeling of the "old" internet so many people are nostalgic for was really just a byproduct of a few enthusiasts gaining an outsize following because the net was financed by universities, governments and the military. Once the unwashed masses and commercial interests joined the party things "went downhill".

You can still blog if you want, you just won't get the automatic exposure.

The nostalgia I feel is for the experimental DIY vibe and doing stuff for free just because it's fun or you want to contribute to a community. Hustle culture and corporate walled gardens have sucked the fun out of the old web.

You can find that same spirit in the discords playing with LLMS right now even though it's walled. It would be better if it wasn't but the spirit is the same.

Got any server recs?

What I miss about the old internet was the lack of commercialization. Commercialization has a habit of making things worse.

Sure but my main point is that the old internet is still there but it's less glamorous. In fact, the old elites benefit from a trickle up effect where the infrastructure is now orders of magnitude better due to the commercial world building out the networks.

> Sure but my main point is that the old internet is still there but it's less glamorous

Some of it is, yes, but a whole lot is gone now. Maybe, perhaps equally likely, it's just very hard to find. The signal-to-noise ratio on the internet is extremely poor.

This is where I settled. The net was kind of hard to use so there was a certain kind of user.

Writing blog posts is kind of hard so there weren’t a ton.

Then it got easy and everyone is on. So thus the yelling and marginalia.

Old internet: a local bookstore/coffee shop near the university housing. New Internet: Grocery store in a big city check out aisle tabloid section(and also the bookstore is still there too)

For blogging to form communities (again'-ish), we're going to need new protocols. I don't think http is the future for growth. The nostalgic internet was a "community," but more because it was a scene made up of other scenes, which each had bars to entry, signalling costs, social proof, status and competence hierarchies, and ultimately, real social networks. Imo, the old internet never disappeared, it just migrated to git.

What we're probably overlooking is that any activity where people can collaborate to make something together at scale is basically magic. I don't think people will go "back" to blogs because the form will need another zero-to-one moment where people collaborate with words to discover something new.

I don't have any use why http would be the barrier. Works for matrix, works for ActivityPub, works for seemingly whatever.

I was really looking forward to http2 letting servers push resources to open connections, but alas the browser implementers have yet to get around to extending fetch to enable push observability. I love the idea of new chat messages coming in as http resourced the page just sees. But even without that nice to have, I want real contentions and real argumentation around how or where http isn't enough. It seems so greatly malleable, can be adapted to serve seemingly any scheme. Looking at grpc for example, it's kind of stunning how little has to be added to make a bidirectional streaming rpc system.

The issue with http is it's interdicted everywhere, requires DNS, TCP, is stateful, and it's tied to a client (browser) that is completely captured and represents an experience controlled by a few organizations. I don't know that streaming is the way for the kinds of relationships and scenes that actually grow.

Https is not interdicted. UDP quic and http3 are stateless. Http is used myriad places outside the browser. You can send whatever you want over http. I hate it but Flutter is showing you can use the web as a low level platform & render however you want, disregarding the standards processes (which are open & democratic and participative, imo).

There is no subculture I would trust a website to join today. Thanks to capture on mobile platforms, the web is homogenized. Few remember that a number of twitter competitors all got banned from mobile app stores under the pretext of potential racism and a real fear of breaking up an oligopoly. The history ISP and corporate proxies and caches would disagree with the statement https isn't interdicted, albeit TLS 1.3 has mitigated that for the most part, but there are still a variety of network security tools that require installing a root cert. Browsers are overcomplicated for anything new to grow out of them, and mobile is a walled garden. New culture will need new tech.

If Twitter competitors kept getting their apps banned, that, to me, would stress the need for the web. Which mobile platforms don't control and block. These sites would have been fine fine if they had been PWAs. The web continues to be the best chance to escape the walled garden.

I agree we need new tech, but those standards and protocols can grow atop http and yes IMHO html too.

> it just migrated to git

It migrated to a version control program? I don't understand what you mean here...

All the valuable collaboration and ideas on the internet occur in git. The rest is entertainment.

I still don't understand. Are you talking about github rather than git?

If so, I don't agree. I don't use github (or any of the other git-based collab sites) but do a lot of collaboration and am exposed to lots of ideas.

If I'm interpreting correctly, you don't understand but you are sure you don't agree, and you do some kind of collaboration without an example and say you are exposed to a lot of ideas without examples either.

I'm sorry, what is the substance of your comment?

The problem is that it is not easy to find a real sense of community because there is no worthwhile way to search the available communities. I find this hugely frustrating. For all their distasteful aspects Twitter and Facebook do help some in this. Perhaps the solution is some sort of dynamic "meta-social medium" for finding communities. If well managed, the economics of such a medium would be huge.

SF writers are hugely creative. I have often thought that a good way to find an idea for a startup would be to read lots of SF then try to implement what I found. Maybe Neil Gaiman will write a story describing some solution to this problem. With his audience, that might be all it would take to spur the creation of such a solution.

Back when Google had honest search, you could find any blog :)

LemmyBB is the kind of direction I wish this would go.

Self-hosted communities, interacting in federated spaces together, specifically in the form of a PhpBB board which I maintain is the least bad form of social media we've tried.

The issue I see with Lemmy, is that you get the Reddit power mod issue, only worse. Add to that the extreme centralization of the instances, both for users, but worse for the hosting of communities, and I can only see dark.

That several instances are openly fascist friendly doesn't help either.

>If well managed, the economics of such a medium would be huge.

This is exactly the problem. Economies of scale ruin communities. Discovery ruins communities. Caring about engagement and metrics ruins communities. Maybe discovery should be difficult, maybe it should take effort.

This also brings to mind the concept of federated search. Frequently brought up as a thought experiment, but it's a shame it's not discussed more seriously.

Something similar to Reddit/lemmy, but without the front page could be interesting.

You have to join the subs and interact with each sub one at a time.

I used Reddit for many years.

I don't remember ever using the "front page".

I always used my own feed where I saw content only from the subs that I followed.

Back before I ditched Reddit (years ago), this was how I used it as well. The "front page" offered nothing of interest to me in that forum.

I used the front page for like 2 days before I realized it was the same content recycled between subs.

No, blogging won't return. Not in the same way as before. In the days of the dial up modem the written word was king. Twitter originally had SMS as a communication mechanism. But look at what the younger generations do, the use the internet for 3D gaming, for video watching.

It's the same micro-attention span as a tweet, but the media has changed. In the immediate term it's video content. But this will also give way to the next thing...

But, unfortunately the written word requires more attention than modern internet users have to spare.

There's still the same amount of intelligent, thoughtful, people with attention spans out there, you could still write for them, the same as always. You just wouldn't be writing for the masses anymore. But do you need to? Do you need your content to be applicable to everyone?

The large platforms enable millions of eyeballs to be aggregated together to attract advertising dollars. In return, the eyeballs receive an algorithmic feed of bite sized dopamine hits (a Tweet, a short video, an Instagram image, etc). It is easier to attract an audience, is easier to sell ads against this audience, but is this audience as valuable to the creator as Cal's 80K newsletter followers? They're different business models and each has its own place.

https://en.m.wikipedia.org/wiki/Eternal_September

1993 was a long time before social media-mediated collapse. Geocities, phpBB, 2channel, and the big IRC nets all postdate Eternal September.

Am I being too judgemental, that Cal did not put a working link to Neil's blog in the post?

Reply on news.ycombinator.com