Notes ·
Atom Is Better Than RSS in Ways That Actually Matter
Chris Morgan makes a convincing technical case for choosing Atom over RSS, not because Atom is newer or more elegant, but because it removes ambiguities that can prevent legitimate content from being represented reliably.
RSS is not really one consistently defined format. The name has been applied to several competing and sometimes incompatible specifications, leaving feed readers to compensate with forgiving parsers, assumptions and implementation-specific behavior.
That usually works, until it does not.
Morgan’s clearest example is the title of a post containing characters such as < or &, or markup such as <code>. RSS does not clearly define whether a title contains plain text or HTML, so different readers can interpret the same title differently.
Atom explicitly defines titles as text constructs. A publisher can declare them as plain text, escaped HTML or structured XHTML, and a conforming reader knows exactly how to process each version. This behavior is defined by the Atom specification rather than left to convention.
Atom also provides a defined distinction between an entry’s summary and its complete content. RSS publishers commonly use description for a summary and the content:encoded extension for the full article, but that arrangement depends on conventions and reader heuristics rather than a clean model built into RSS itself.
That distinction matters when building software intended to exchange content reliably. A format should not require every application to guess whether a field contains text, HTML, a summary or an entire article.
The unfortunate exception is podcasting. Most ordinary feed readers and parsing libraries support Atom, but major podcast platforms remain centered on RSS. Morgan traces much of that history to Apple supporting RSS for the iTunes podcast directory during the period when podcast distribution standards were becoming established.
RSS remains dominant largely because software has become very good at tolerating its inconsistencies. That does not make those inconsistencies desirable.
Atom was designed to solve these problems and standardized as RFC 4287 in 2005. Twenty years later, developers are still building new systems around the format Atom was intended to replace.
Calling every web feed “RSS” also hides an important distinction. RSS may be the familiar name, but Atom is the better-defined format for exchanging articles, notes and other structured web content without relying on guesses.
The article is currently marked as a draft and Morgan plans to expand and test parts of the argument. Even unfinished, it offers one of the clearest explanations I have read of why the differences between RSS and Atom are not merely technical trivia.