Dave Winer froze RSS 2.0 and gave its copyright to Harvard — and the format won

Netscape created RSS in 1999 to feed its My.Netscape portal, then abandoned it: in April 2001 new owner AOL removed RSS support and deleted the documentation. Two camps filled the void. The RSS-DEV Working Group — Aaron Swartz, O'Reilly, Moreover — built RSS 1.0, a cleaner design that restored RDF and namespaces. Dave Winer's UserLand, which had shipped the first RSS publishing tools outside Netscape, continued the simple 0.91 line. Neither party could claim the format's name; UserLand's trademark filing died in December 2001, and the community split over which RSS was real.

In September 2002 Winer released RSS 2.0 — Really Simple Syndication — a backward-compatible step from 0.92. Then came the unusual move: in July 2003 Winer and UserLand assigned the copyright of the RSS 2.0 specification to Harvard's Berkman Center, and spun up an RSS Advisory Board to publish it. Winer announced the handover on scripting.com that month (July 28), answering one of the major objections to RSS. The spec states it plainly: 'the RSS spec is, for all practical purposes, frozen at version 2.0.1' — future work belongs in modules, namespaces, or new formats.

The counterpart, Atom, began in June 2003 and took the committee route: an IETF Proposed Standard only in 2005 (RFC 4287). The frozen format needed no such patience. Between 2005 and 2006 RSS went mainstream — Microsoft's IE and Outlook teams adopted Firefox's orange feed icon in December 2005, Opera followed in February 2006, and browsers made subscribing a one-click default (Wikipedia). The last revision of record, RSS 2.0.11, is dated March 30, 2009; seventeen years on, nothing has changed, and podcasting still ships over the format Winer froze.

Freeze the spec and give the copyright away.

Freezing removed the waiting cost: a publisher plugging into RSS 2.0 in 2003 was implementing the final version, not betting on a roadmap.

Giving the copyright to Harvard removed the toll and the villain: under a Berkman-hosted license, no opponent could argue the format was one company's property.

With no versions left to want, the RSS-DEV argument about a cleaner 1.0 lost its audience — 2.0 ended the debate by refusing to change.

The two-camp fight itself marketed feeds: years of RSS-against-RSS coverage taught web publishers that a feed was something a modern site simply had.

Browsers needed stability more than elegance: an orange icon and one-click subscribe only made engineering sense against a format guaranteed not to move.

A spec can win by ending: ship something good enough, freeze it, hand away the keys. Ownership invites rivalry; a finished common good invites adoption — rivals fight the absence of a decision.

Atom became the IETF's answer and never displaced the thing it answered: the two formats coexisted through the aggregator era, with RSS 2.0 reaching mass use in 2005–06 as browsers built it in. The frozen-spec discipline held even in governance — Rogers Cadenhead relaunched the RSS Advisory Board without Winer in January 2006, and its 2007 revision only clarified. RSS 2.0.11 (2009) remains the last revision. Google Reader's 2013 shutdown drained the reader market, but the enclosure tag from 2000 still carries every podcast, and the frozen spec outlived the war it was frozen to end.