When Google shut down Google Reader on July 1, 2013, a lot of people decided RSS had died with it. I can understand why. For many people, Google Reader was RSS, and when the most visible way of using a technology disappears, it’s easy to assume the technology itself went away.
RSS didn’t go anywhere. The best example is podcasting.
A podcast is, fundamentally, an RSS feed. Your podcast app checks a feed on somebody’s server, sees whether there’s a new episode, and follows a link to the audio file. There are more layers around it now, of course, but the basic mechanism is still remarkably simple, and that simplicity matters.
Podcasting became a large commercial business without requiring everyone to use the same company’s network. A podcaster can publish a feed on one server, and listeners can subscribe with any compatible application. Apple doesn’t own the feed. Spotify doesn’t own it. Neither does Google.
That’s one reason podcasting has remained relatively open despite repeated attempts to turn it into another collection of proprietary platforms. Companies can build services around RSS, but they don’t own RSS itself. A standard is much harder to lock up than a product.
RSS is still underneath a lot of the web
Plenty of websites still publish feeds, whether their owners think about them or not. WordPress creates feeds by default. Substack newsletters have them. YouTube channels have feeds. Reddit exposes RSS for many pages, and GitHub publishes Atom feeds for things such as releases and commits.
None of this required an RSS revival or some announcement that feeds were back. They simply never left.
Part of the reason RSS seems invisible is that it isn’t a product. There’s no RSS corporation selling subscriptions, running advertising campaigns, or trying to convince investors that XML files are the next big thing. It’s infrastructure, which means that when it works, most people don’t have much reason to think about it.
SMTP has much the same problem. Email still depends on standards developed decades ago, but nobody writes breathless articles announcing the return of SMTP. It simply carries mail. RSS is like that. It’s old enough, boring enough, and useful enough that it mostly disappears into the plumbing.
Google said Reader was being shut down because usage had declined, and I have no particular reason to doubt that Reader itself had declined. But the timing is still interesting. Google was heavily invested in Google+ at the time, and Reader represented a very different idea of how people might use the web.
With RSS, you chose the sources and the reader, and you decided what appeared in front of you. That model wasn’t especially useful to a company trying to build another social platform.
The case for RSS has gotten better
A feed reader does something almost embarrassingly simple: it shows you the things you subscribed to, usually in chronological order. That’s about it.
There’s no algorithm deciding that a stranger’s argument is more important than the writer you deliberately followed. There are no promoted posts inserted because somebody paid to be there, and there’s no need to keep scrolling while a recommendation system tries to guess what might hold your attention for another thirty seconds.
If a site publishes twice a year, you get two items a year. If it publishes nothing, you get nothing. That sounds almost primitive compared with modern social media, and I mean that as a compliment.
A good feed reader is boring, and boring software is underrated.
Over the past decade, social timelines have become increasingly shaped by ranking systems, advertising, recommendation engines, and whatever metric a platform happens to be optimizing this quarter. RSS readers, meanwhile, kept doing what they always did. The software improved, too.
Feedly and Inoreader handle the hosted side of things. NetNewsWire is a good free option for Apple users. FreshRSS and Miniflux work well if you prefer to run the service yourself. None of these requires much explanation. You subscribe to feeds, new items appear, and you read them. That’s enough.
RSS has limitations
There’s a reason RSS never became the default way most people use the internet. You have to know what you want.
A new feed reader is an empty screen. TikTok, YouTube, Facebook, and similar services immediately start offering things to watch or read. Their recommendation systems may be intrusive, manipulative, useful, irritating, or all four at once, but they solve a real problem: discovery. RSS has never been especially good at that.
There’s also the unread counter. Anyone who has opened a feed reader and discovered 1,400 unread items knows how quickly a tool for reading can start looking like unpaid clerical work. Eventually you stop opening it because the number keeps getting worse. That isn’t a small problem, and it’s one of the reasons people abandon feed readers.
But RSS doesn’t need mass adoption, and that may be one of its strengths.
There’s no central company that needs continued user growth to justify keeping the service alive. Nobody has to hit an engagement target. Nobody can buy RSS and announce six months later that the free tier is being discontinued. It’s just a published format that other software knows how to read, which makes it difficult to kill.
RSS has survived because it is cheap, simple, decentralized, and useful. Those aren’t exciting qualities, but they tend to be good qualities for infrastructure.
If you regularly read a website, there’s a decent chance it still has a feed somewhere. Sometimes it’s obvious. Sometimes you have to look for /feed, /rss, or an RSS link buried in the page source.
The web never stopped publishing feeds. Most people just stopped looking for them.