The Small Web Still Depends on Corporate Infrastructure

I started thinking about this while trying to join a few webrings that actually matched my website. The first one I joined was the No AI Webring, and surprisingly, the owner approved my site in less than an hour.

After that, I started exploring other webrings. Yesterweb immediately interested me because its philosophy aligned strongly with my own views on the modern internet, but the project has effectively been discontinued. Then I looked at the Bucketfish Webring, only to realize that the only way to join was by submitting a pull request on GitHub. The same thing happened again with the XXIIVV Webring. In both cases, participation in supposedly independent web communities still depended on having an account on a centralized Microsoft-owned platform.

The contradiction started becoming difficult to ignore. Many parts of the so-called “independent web” still rely heavily on centralized infrastructure somewhere in the stack. GitHub has quietly become a gatekeeper for participation in communities that are often explicitly built around resisting centralization.

I then looked at larger communities like IndieWeb and Hotline Webring, but they introduced a different problem entirely. At a certain scale, a webring stops feeling like a ring and starts feeling like another directory. The original purpose of a webring was not infinite expansion. It was supposed to be a small collection of interconnected websites where people could realistically discover each other through actual browsing instead of algorithms. Once a webring becomes too large, the sense of connection weakens and the ring itself starts functioning more like search results than a community.

The Small Web Is Not Fully Independent

There is a comforting story people like to tell about the small web. It says that if enough of us care about independence, we can build something cleaner, freer, and more human than the corporate internet. We can make our own sites, run our own servers, avoid the big platforms, and slowly recover a web that belongs to people again.

I sympathize with that story. I still want it to be true. But the reality is less elegant.

The small web is not fully independent. Not really. It may reject platform culture in spirit, it may look decentralized on the surface, but it still leans heavily on corporate infrastructure underneath. GitHub hosts code, Neocities hosts sites, Discord and Reddit circulate links, Cloudflare sits in front of pages, Google Fonts appears in the background, and the wider habits of the internet still pull people toward centralized spaces even when the content itself is personal, handmade, and sincere.

That does not mean the small web is fake. It means it is living inside a system it did not build.

The Tradeoff Between Independence and Reach

Hosting is not enough. A website is not only files on a server. It is also a path into attention. It needs circulation. It needs discovery. It needs repeated exposure. A page that exists but cannot be found is not functionally the same as a page that can be shared, indexed, recommended, and stumbled upon by strangers. Corporate platforms understand this very well. They do not merely host content; they arrange the conditions under which content becomes visible.

That is why people stay.

The question is usually framed too simply. It is easy to say that people rely on corporate platforms because they are lazy, or because they have no principles, or because they are too comfortable to leave. Sometimes that is true in a shallow sense, but it is not the real explanation. The real explanation is incentive structure. Corporate infrastructure lowers friction in the places that matter most.

It gives users the most other users. It gives them the easiest onboarding. It gives them trust by default. It gives them immediate participation instead of setup. It gives them the strongest discovery layer, even when the content itself is mediocre. It gives them all of these things at once, which is why it remains hard to replace.

People often want independence, but they also want readers. They want ownership, but they also want reach. They want decentralization, but they also want a place where people actually show up. They want to avoid platform dependence, but platform dependence is often what makes discovery possible in the first place. The contradiction is not a sign of stupidity. It is a sign that the internet has become socially organized around systems that are difficult to replicate on a smaller scale.

That is why the small web keeps borrowing from corporate infrastructure even when it dislikes doing so. A site can be beautiful, personal, and fully controlled by its author, and still fail to matter socially if it has no way to circulate. Technical independence does not automatically produce cultural presence. The web is not simply a network of servers.

It is a network of habits, referrals, and shared attention.

That part of the system is harder to decentralize than the code.

This is why Neocities remains useful even to people who would prefer something else. I do not mean useful in an abstract ideological sense. I mean useful in the most ordinary way: it gets people seen. It gives small sites a discovery layer that isolated hosting often cannot provide. A personal server may be more independent, but it is often much harder to stumble into. Neocities is a compromise, and that word should not be treated as an insult. Most real web use is compromise. The question is not whether a tool is pure. The question is whether it makes a meaningful life online possible.

This is the part that makes the small web harder to talk about honestly. People often want to turn it into a purity contest. They want to say that if you use GitHub, or Discord, or Neocities, or Cloudflare, then your commitment to independence is fake. But that is too simple. It ignores the actual cost of refusal.

Refusing corporate infrastructure is not free. It usually means more setup, less audience, more maintenance, fewer contributors, and slower social life. It means asking people to cross extra steps just to reach you. It means accepting that your work may be harder to discover, harder to use, and harder to sustain. Those costs are not imaginary. They are immediate. And when people choose the easier path, that choice is often rational rather than weak.

That is why the accusation of hypocrisy misses the deeper issue. The problem is not merely that people say they value independence while using centralized tools. The problem is that the alternatives usually do not match the social function of the tools they reject. If the only way to preserve independence is to accept invisibility, then many people will not do it. That is not because they love corporate systems. It is because human beings are social, and social existence has requirements.

A fully self-hosted site can be technically elegant and still feel lonely. In fact, loneliness is often the hidden price of independence. You can own your page, your domain, your server, your files, your design, and your data, but if nobody sees it, those victories become private. They matter, but they do not create communities by themselves. A web made of isolated personal islands may be morally satisfying, but it does not automatically produce a public.

A person can reject every centralized service and still end up alone. Sometimes that is a principled choice, but sometimes it becomes a form of disappearance. Independence without reach can become self-erasure. I do not say that to discourage people from taking control of their own work. I say it because I think the cost needs to be named clearly. The internet rewards visibility, and visibility is still concentrated in places that do not share the values of the small web.

Why I Use Neocities

I use Neocities partly because I dislike the idea of the web collapsing into a handful of companies, but also because self-hosting is simply not practical for me right now. That is the honest answer. I do not use Neocities because I believe centralization is good. I use it because it is the best compromise currently available to me.

What Neocities provides is not just hosting, but discoverability. Neocities still has a browsing culture. People explore profiles, follow links, read guestbooks, exchange buttons, and occasionally discover a site by accident. I know this because it already happened to me. Someone found my website on their own, read enough of it to care, and linked my button on their own site. That kind of discovery is difficult to reproduce on an isolated VPS with no surrounding ecosystem.

This is the contradiction people rarely say out loud when discussing the “independent web.” Total independence often comes at the cost of visibility, while visibility usually depends on some form of shared infrastructure. Even many anti-corporate web communities still rely on centralized platforms somewhere in the stack, whether that is GitHub, Discord, Cloudflare, or something else. Neocities is centralized too, but unlike most platforms, it still encourages outward linking instead of trapping people inside itself.

So for now, I treat Neocities less as a permanent home and more as a bridge. If I eventually have the resources to self-host properly, I would still likely keep a presence here, simply because it remains one of the few places on the modern web where independent websites can still organically find each other.

Centralization Solves Real Problems

So when people remain on GitHub, Neocities, Reddit, Discord, or similar platforms, the explanation is often simpler than critics want it to be. They are not necessarily endorsing the system. They are adapting to it. They are choosing the route with the least resistance. They are choosing the place where the social gravity already exists. That gravity is not imaginary. It is the reason a repository gets contributors, a site gets readers, and a project gets remembered.

In practice, centralization solves real social problems. It reduces uncertainty. It lowers the cost of participation. It makes sharing easier. It gives people confidence that their time will not be wasted reaching an empty room. This is a powerful advantage. Alternatives have not yet matched it at scale, which is why the same corporate systems continue to mediate so much of the supposedly independent web.

Centralized platforms become harder to leave because their value compounds socially. People stay where the links already point, where contributors already gather, where audiences already exist, and where habits have already formed. Even communities that dislike corporate infrastructure often rebuild themselves around the same gravitational centers because starting over socially is harder than starting over technically.

The Real Challenge Is Rebuilding the Social Layer

The small web is not dead, but it is still standing on borrowed infrastructure. Its values may be better than the systems it depends on. Its aesthetics may be more humane. Its culture may be more thoughtful. But those things do not eliminate the structural fact that the wider web still routes attention through centralized channels. A handmade site does not become socially autonomous just because it is handmade.

This is why it is misleading to treat the small web as a simple opposition to the corporate web. The two are not cleanly separated. They overlap. The small web often survives by using the corporate web as a transmission layer. It may reject platform logic in its style, but it still needs platform gravity in order to be noticed. That dependency is not always visible from the outside, but it is there in the everyday work of publishing, linking, and hoping to be found.

I do not think this should produce cynicism. It should produce clarity.

It is easy to romanticize independence until you have to maintain it. It is easy to praise decentralization until you notice how much extra effort it demands from ordinary users. It is easy to demand that everyone leave the big platforms, but harder to build replacements that are easy enough, useful enough, and socially dense enough to matter. The internet did not become centralized only because people were careless. It became centralized because centralization solved problems that people cared about immediately. Convenience won not because people are shallow, but because convenience is a real force in human life.

That does not mean the corporate model is the final answer. It only means the alternative has to do more than appeal to principle. It has to offer a social environment people can actually live in. It has to make discovery easier, not harder. It has to reduce the cost of participation instead of increasing it. It has to recreate the pathways that let a page become public, not merely existent.

This is the real challenge for the small web. Not just hosting your own pages. Not just avoiding corporate branding. Not just writing clean HTML or owning a domain. The harder task is rebuilding the social layer that lets people find one another without handing all discovery to a few large companies. Until that happens, corporate infrastructure will remain the scaffolding beneath much of the supposedly independent web.

The small web is alive, but it is not self-sustaining in the way its supporters hope. It still borrows trust, circulation, and attention from larger systems. That is not a failure of character. It is a structural condition. The people who use centralized tools are often making the most practical decision available to them, not the most ideologically comfortable one.

And that, to me, is the most honest way to understand the situation. Not as betrayal. Not as surrender. Just as dependence shaped by incentives.

The small web still matters. It still offers something that the big platforms cannot: personal voice, slower pace, deliberate design, and a sense that the page belongs to a person rather than to a machine. But it remains entangled with the infrastructure it would rather not need. Until discovery itself becomes easier to decentralize, that dependence will continue.

The task is not only to build pages. It is to rebuild the conditions under which those pages can be found. Until then, the small web will keep living on corporate infrastructure, even when it does not want to.