RSS isn’t dead, just quiet

Netscape released RSS in 1999, as a format for pulling other sites’ headlines onto your own start page. A few years later it was everywhere: blogs had an orange icon in the corner, news sites offered dozens of feeds, and for many people Google Reader was the first tab of the morning. Then social networks arrived with their timelines, Google switched its Reader off in 2013, and RSS disappeared from most people’s view.

To be honest, RSS is not a mass phenomenon today. Most people follow content through LinkedIn, Instagram or a newsletter, and many do not even know there is an alternative. Yet RSS has not gone away, it has just gone quiet. Every podcast you listen to is technically an RSS feed that your podcast app fetches at regular intervals. And for anyone who wants to subscribe to content without paying for it, namely with their data, it is still one of the most elegant solutions on the web.

How a feed works

At its core, a feed is a simple XML file at a fixed address. It contains the latest entries of a website, each with a title, a link, a date, a unique identifier and a description. That is all. There is no server actively sending anything, and no list of subscribers. Instead, a program, the feed reader, fetches the file at regular intervals and shows what has been added since last time.

<item>
  <title>RSS isn’t dead, just quiet</title>
  <link>https://example.com/articles/rss</link>
  <guid isPermaLink="true">https://example.com/articles/rss</guid>
  <pubDate>Wed, 30 Sep 2026 00:00:00 GMT</pubDate>
  <description>A standard from 1999 …</description>
</item>

So that nobody has to hunt for the feed’s address, a website announces it in the head of every page. This so-called autodiscovery is a single line, and it means that in most readers it is enough to paste the ordinary address of a website. The reader then finds the feed on its own.

<link rel="alternate" type="application/rss+xml"
      title="Compendium" href="/compendium/rss.xml">

Besides RSS 2.0 there is the Atom format, described as an official internet standard in 2005. It is defined a little more strictly but follows exactly the same principle, and every serious reader understands both. For readers the difference hardly matters.

Privacy by design

This is where RSS really shines. Subscribing to a feed needs no account, no email address and no consent to anything. The feed sets no cookies and runs no JavaScript, and the website’s operator does not know who reads their content. There simply is no subscriber list that could be analysed, sold or leaked. Unsubscribing means deleting the entry in your reader, with no unsubscribe link and no question whether you might want to reconsider.

To complete the picture, two honest caveats. When the feed is fetched, the server sees an IP address, as with any request on the web. If you use a cloud service as your reader, though, the website operator does not even get that, because the service fetches the feed once for all of its users. And technically, an operator could embed an invisible counting pixel in a full-text feed, as many newsletters do. A decent feed simply leaves that out, and many readers only load external images on request anyway.

Personalised, but by the reader

Social networks personalise too, but for the platform. An algorithm decides what you see, in which order and how often, optimising for time spent rather than for your time. With RSS it is the other way round. You decide which sources to subscribe to, how to sort them into folders, whether to read by date or by source, and above all, when to read. There is no notification that pulls you out of your work, and no post that sits at the top because it generates a lot of outrage. Nothing gets lost because the algorithm thought it unimportant.

RSS is also personal when it comes to whom you trust with your subscriptions. There are three kinds of reader. Local apps run directly on your device, keep subscriptions and reading state there and fetch the feeds themselves, with no account at all. These include NetNewsWire for Mac and iPhone, which is free and open source, Reeder for Apple devices, Feeder for Android, and the built-in feed readers in Thunderbird and the Vivaldi browser. This is the most data-frugal option, because nobody but you knows what you read.

Cloud services such as Feedly, Inoreader, NewsBlur or Feedbin store your subscriptions with them and sync them across all your devices. That is convenient, but the provider knows your subscriptions, while the websites themselves only see the service, not you. The third option is a self-hosted reader such as FreshRSS or Miniflux. Both are open-source software that you install on your own server and then use in the browser. Subscriptions and reading state stay with you and are still the same on every device, and the feeds are fetched by your server, not by your phone. That takes some technical know-how, but for teams and tech enthusiasts it is an elegant solution. By the way, the three kinds are not mutually exclusive, since many local apps can connect to a cloud service or your own server.

Preview or full text

For each entry, a feed can deliver just a short description or the complete article. In RSS 2.0 the preview goes into the "description" field, the full text usually into a separate field from a small extension, "content:encoded". Atom makes a very similar distinction between "summary" and "content".

Both have their place. A full-text feed is the most convenient for readers, because you can read the whole article in the reader, even offline on the train. A preview feed is smaller, loads faster, and brings interested readers to the actual page, where there are graphics, code examples, a calculator or a way to get in touch that a reader cannot display. For our compendium we chose the preview, because many articles live on exactly these elements. If you publish texts that stand on their own, such as a blog or news, your readers will thank you for the full text.

What makes a good feed

A feed is quickly generated; a good feed needs a few more details. Every entry gets an identifier that never changes, because if it changes, the reader shows the same article as new again. The date has to be correct and in the prescribed format, or readers sort things wrongly. Links are always complete addresses including the domain, because a reader cannot know which site a relative path was meant for. The feed states its language, and a multilingual website is better off offering one feed per language than a mix.

Then there are a few touches that are easy to forget. The feed points to its own address with a self link, so readers know where it lives even after redirects. It is announced via autodiscovery in the head of every page and also linked visibly, for example in the footer, because not everybody knows their reader can find the address by itself. If you set up the server to answer briefly with "nothing new" when the content has not changed, instead of sending the whole file every time, you save bandwidth for yourself and for the readers. And before a feed goes live, a run through the free W3C Feed Validator is worth it.

Feeds for search engines

A feed helps not only people but also search engines. In its sitemaps section, Google Search Console accepts RSS and Atom feeds alongside the classic sitemap.xml, and Google explicitly recommends using both. The sitemap is the complete directory of all pages and is re-read rather rarely. The feed contains only the latest content, is small, and is therefore fetched much more often. Why this combination helps especially when moving your website is described in our article Moving house without losing your reach.

What this means for you

RSS will not make a comeback as a mass medium, and it does not need to. It is a small, open standard that has worked for more than 25 years, collects no data and leaves readers in control instead of handing them to an algorithm. For a website, a feed costs almost nothing: a file generated from the same data as the pages themselves, one line in the head and a link in the footer. In return you get a channel that belongs to nobody but you, and a faster line to the search engines. And if you want to try how that feels, our compendium is a good place to start.