There was a time when going online usually meant visiting a website. It might have belonged to a newspaper, a university, a company, a hobbyist, or someone who had spent an unreasonable amount of time documenting old computers, local history, ham radio equipment, or anything else that interested him. Some of those sites were well designed and some were awful. There were blinking graphics, unreadable backgrounds, bad color combinations, visitor counters and more “Under Construction” signs than anyone needed. Still, they belonged to the people who made them.
That seems like a small distinction until you look at much of the Internet today. We still use websites, of course, but more of our online lives take place inside platforms. Facebook, Instagram, YouTube, Reddit, TikTok and similar services give us a place to publish, communicate and find an audience without having to know anything about web hosting, HTML, databases or servers. That convenience solved real problems, and I am not interested in pretending the Internet of 1998 was some sort of technological paradise.
It wasn’t.
But in making the Internet easier, we also gave up a surprising amount of control.
From Places to Accounts
One of the strengths of the early web was that a site had an address. You could bookmark it, link to it, send the address to somebody else, or come back a year later and hope it was still there. The site might move to another server or change hosting companies, but if the owner kept the domain name, most visitors never had to know.
That arrangement put a fair amount of control in the hands of the site owner. A person who owned a domain could move from one host to another. Software could be changed. The design could be replaced. The entire site could be backed up and restored somewhere else. None of this guaranteed permanence because websites disappeared then just as they disappear now, but the owner had choices.
Platforms changed that relationship. Instead of building a place on the web, we increasingly created an account inside somebody else’s system. Our identity became tied to that company, along with our audience, photographs, posts, conversations and everything else we had accumulated there. Moving no longer meant copying files to another server. It meant starting over somewhere else, assuming the service even gave you a practical way to take your material with you.
We gradually became accustomed to renting our online identity.
That arrangement works fine until the landlord changes the rules.
A company can change how your material is displayed, decide who sees it, restrict outside links, alter its terms of service, suspend an account, discontinue an API or shut down the service entirely. None of those things are hypothetical. They are ordinary parts of doing business on a privately owned service.
With a personal website, I may still depend on a hosting company, a registrar and an Internet connection, but I can replace those pieces. If my hosting company becomes expensive or annoying, I can move. The ability to leave is an important form of control.
The Internet Used to Depend More on Standards
Another thing we lost was a certain amount of interoperability.
A great deal of the older Internet was built around protocols rather than membership in a particular company. Email is probably the best surviving example. I do not need to use the same email provider you use. Someone with a Gmail account can send mail to someone using Outlook, Fastmail or a server running in a basement. The software can be completely different because both sides agree on how email is exchanged.
The same idea applied to other parts of the Internet. Usenet had different readers. IRC had different clients. FTP had different programs. The web had different browsers and servers. RSS eventually allowed people to follow websites using any compatible reader they preferred.
None of these systems were perfect. Usenet became overrun with spam. IRC could be confusing. Running your own mail server today can be an exercise in discovering just how much large providers distrust everyone else. Open systems develop their own problems, and sometimes centralized services genuinely do a better job of handling them.
Still, interoperability gave users choices.
Imagine if email worked like many social networks. Gmail users could only send messages to other Gmail users. Outlook users could only communicate with Outlook users. Changing providers would mean abandoning everyone you had communicated with and hoping they followed you to the new service.
We would consider that absurd, yet we accept a similar arrangement from most social platforms.
There are economic reasons for that. A service becomes more valuable when everyone you know is already there, and leaving becomes more difficult for the same reason. A company has little incentive to make it easy for customers to take their relationships somewhere else. Open protocols weaken that control because they allow competing software and services to communicate.
For users, that weakness is actually useful.
Somebody Else Now Decides What We See
The change from websites to platforms also changed how information reaches us.
A traditional website is fairly straightforward. The owner publishes something and the reader visits the site. RSS made the process even simpler by letting readers subscribe to updates from many independent sites and read them in one place. The relationship remained understandable. The author decided what to publish, and the reader decided what to follow.
Platforms inserted another party into that relationship.
Once millions of people began posting material in the same place, chronological order became difficult to manage. A person might follow hundreds or thousands of accounts, producing far more material than anyone could reasonably read. Platforms responded with recommendation systems designed to decide what a user was most likely to watch, read or react to.
There is nothing inherently sinister about sorting information. Libraries do it. Newspapers do it. Search engines do it. The difference is that a platform usually has business goals connected to the sorting.
If a company makes money by keeping people on the service, it has a reason to show material that keeps people scrolling. If advertising pays the bills, the system has a reason to produce more opportunities to display advertising. If reactions and shares keep people returning, controversial or emotionally charged material may perform better than something thoughtful that takes ten minutes to read.
The result is that we have less direct control over our own reading.
With RSS, I decide which sites are in the list. When a site publishes something, it appears. Nobody decides that a three-day-old argument deserves priority over a new post because it might keep me staring at the screen longer.
That difference is easy to overlook because algorithmic feeds are convenient. They also place a private company between the writer and the reader.
The Web Became More Uniform
Personal websites used to be strange.
That was frequently a bad thing. Anyone who spent time online in the 1990s remembers pages that looked as though every available HTML feature had been used simply because it existed. Modern web design is usually cleaner, more readable and more accessible. Content management systems have made publishing much easier, and there is nothing wrong with using a decent template instead of designing every page from scratch.
Something was lost when almost everyone moved into the same handful of systems, though.
A personal website could reflect the interests and personality of the person running it. The structure itself could tell you something about the owner. One site might be a collection of essays. Another might be a giant directory of links. Someone else might maintain a technical notebook, a photo archive or a collection of scanned manuals for hardware nobody had manufactured in thirty years.
Platforms tend to flatten those differences. Everyone gets roughly the same profile page, the same posting tools and the same limited set of ways to organize information. You supply the material, but the company decides what a page looks like and how people move through it.
That uniformity makes services easier to use. It also makes much of the modern web feel strangely repetitive.
The old web could be ugly, confusing and badly organized. It was also full of places that clearly existed because one particular person cared enough about something to make a site about it.
I miss that more than I miss animated GIFs.
Platforms Solved Problems
It is easy to criticize platforms after they became enormous, but there are good reasons people moved toward them.
Running a website used to require more technical knowledge than most people had or wanted. Hosting photographs and video could become expensive. Finding an audience was difficult. Search engines were not always useful. Software had to be installed, maintained and patched. Spam, security problems and server failures were the responsibility of whoever ran the site.
A platform removed most of that work.
Someone could create an account and begin publishing in minutes. Families could share photographs without learning how to run a server. Community organizations could communicate without hiring someone to maintain a website. Musicians, artists and writers could reach people who might never have found an independent site.
Those are genuine advantages, and I use centralized services myself when they are useful.
The problem came when convenience turned into dependence and we began treating a few privately owned systems as though they were the Internet itself.
Facebook is not the web. Neither is YouTube, Reddit, TikTok or any other company. They are services that exist on top of the Internet. Keeping that distinction in mind is becoming increasingly important.
The Independent Website Still Has a Purpose
The good news is that most of the pieces that made the independent web possible are still here.
You can still buy a domain. You can still rent inexpensive hosting or run a server yourself. WordPress still exists. Static site generators exist. RSS still works. Email still works. Projects such as Mastodon have shown that social networking does not necessarily require one company controlling the entire system.
None of this means everyone should run a server. Most people probably should not. There is no virtue in making technology more difficult merely to prove that you can.
There is value, however, in having a place online that you control.
A personal website does not need a business plan or thousands of visitors. It can be a collection of writing, notes, photographs, links, technical information or whatever else the owner considers worth keeping. It can change over time without asking permission from a platform operator. It can remain small.
Most importantly, it can connect to other independent sites without requiring all of them to belong to the same company.
That simple idea helped make the web useful in the first place.
Platforms are probably going to remain a major part of the Internet. They are too convenient to disappear, and many of them provide services that would be difficult for individuals to reproduce. I have no interest in abandoning useful technology simply because an older method gave users more control.
I do think we should remember what that control was worth.
Owning a small piece of the web means having somewhere that does not disappear because an account was closed, a recommendation system changed or a company decided your audience was no longer profitable. It means keeping your files, choosing your software and deciding how your material is organized.
The web was built so that millions of independent places could connect to one another.
There is still room for a few of them.