#StandWithUkraine

Russian Aggression Must Stop


Web of sites

2025/02/23

Tags: tech

The web has a centralization problem. A good chunk of most people's usage of the web is focused around massive web platforms, such as mainstream social media sites, operated by mega-corporations. These platforms do a lot of things for their users: they allow users to view and produce so-called "content" and usually provide recommendation and discovery tools to ensure their users receive a constant stream of "content" that the platform has determined the user might be interested in and find engaging. All they ask for in return is your undivided attention and any and all information they can glean from you, so that they can sell that data to advertisers and anybody else that might be willing to buy it.

The web of the big tech

This web 2.0 approach has meant that it's easier than ever for people to both produce web content and consume it. And this approach has no doubt been proven to be highly successful. Technology Connections recently released a video on a phenomenon he named the "algorithmic complacency". This algorithmic complacency manifests itself as a desire to let algorithmic feeds pick the content one watches, as opposed to curating content by yourself.

All of this has given the companies that operate these platforms immense amounts of power. In the best scenario, they are able to gain access to a vast amount of information on their user-base and establish walled gardens, where people are heavily tied into their platform, especially as content creators. For example, if you operate a YouTube channel as your business, ending up in a dispute with YouTube could very well prove fatal to your entire business.

In the worst scenario, this power can manifest itself as a tool for mass-scale opinion modification. It has been generally known that disinformation campaigns have been conducted across the mainstream social media sites especially during election cycles to push extremist propaganda, typically in the form of far-right narratives. In the past this was usually done by foreign troll farms, but especially as the United States slides into a far-right techno-oligarchy, it ought to be expected that these platforms are going to increasingly push narratives that are favorable to themselves. And we've already seen that even when the platforms aren't necessarily actively doing it, these algorithmic feeds and recommendation engines are very capable of pushing people into extremist pipelines, gradually shifting the views of people towards more twisted and realism-detached ideologies.

There are of course alternatives to this. Federated social media, such as the Fediverse and its many projects, like Mastodon, Pixelfed and PeerTube, are a worthwhile attempt to push back against the centralization and alternative forms of personal content curation. Typically they focus more around a chronological timeline and an organic discovery process focused around boosting or sharing content from others. Thus, by following just a few people, you discover more people by looking at who else's content they share and so on. Proper federation also decentralizes power, such that one group of people hopefully aren't able to subvert the system to their direct advantage at the expense of others. I personally use federated social media and the Owncast server I host also plugs into the Fediverse.

Regulatory pressure towards centralization

However, federated web 2.0 platforms by themselves also do carry some risks, although mitigated compared to pure big-tech centralization. These platforms are still acting as the gate-keepers of sorts and there are potential forces which may push user-generated content back towards a more centralized state. A good example of this is the UK government's Online Safety Act which aims to regulate the safety practices of platforms hosting user-generated content, with the usual "think of the kids" concerns. The OSA is considerably broad and involves pretty heavy bureaucratic requirements to comply, leading to a number of forums and other smaller websites to either limit their functionality or shut down entirely. It's worth noting that big platforms have not made a peep about this, since they are likely so heavily lawyered up and/or have more than enough capacity to deal with government bureaucracy that they are effectively unaffected.

So, is there another way to combat centralization such that you are not going to take up a significant burden from regulations like the OSA? Naturally, but we're going to have to go backwards a little bit.

Back to 1.0

You are reading this post on a website that is hosted by myself. Well, technically the hardware it is hosted on is not owned by me, but I pay a monthly fee for a (virtual) server that I can manage. I run a web server on it and I can push new pages of content on it when I please.

How you stumble upon this page is where things get a bit tricky. Possibly the easiest way to get people to go onto some site is to post a link to social media, but that introduces a dependency on either a major big-tech platform or some federated service. Doing so is not a bad thing by any means, but assuming we like in a hypothetical world where we aren't doing user-generated content, that wouldn't be possible.

The good thing is that people figured out ways to generate aggregated feeds of content prior to the popularization of social media. The software I use to generate the HTML files and such for the website also produces an RSS feed. Basically, it's just a list of timestamps, post titles and links that an RSS reader application can pull and interpret to produce a human-readable view of posts from this website. An RSS reader is also not limited to just a feed from one website, but can instead pull the RSS feed from a whole bunch of sites. There are still plenty of RSS readers out there. I use one called Elfeed, but an Emacs plugin is probably not for everyone. Something like Liferea is probably easier to get started with.

That still of course leaves the problem of discoverability. And this is honestly not something I am doing very well at the moment with my website either. However, there is an idea called a blogroll which works basically exactly like how content curation via boosting works on Fediverse. If you find a blog or a website you like, you put up a link to it on your website. This creates a web of blogs cross-linking to each other, organically recommending cool and useful places to check out. And once you've found a site, you can grab its RSS feed link and "subscribe" to the new posts that will appear in the future.

Note to self: I probably ought to set that sort of thing up here.

With a little bit of technical know-how and some disposable income, it's possible for a good chunk of people to do this themselves. Now don't get me wrong, it still takes some effort to figure out how all of it works and even modest hosting costs are likely too much for some people. The practicalities of running your own website and blog are very much dependent on who you are, who your audience is and in what sort of a life situation you find yourself in. But, I think it is worth highlighting, since I think it represents a good bit of agency-per-buck ratio and especially as a guarantee against centralizing effects of government regulation and anti-competitive measures by big-tech.

So, if you are able, go set up a feed reader for yourself and grab a few feeds into it. And if you are able and interested, get yourself a blog. There are plenty of guides and tutorials out there for how to do it, but I'll probably put one together at some point as well.

>> Home