A look back at the evolution of RSS and the reasons it has lost the benefit of enterprise-controlled, web-centric information silos, although their format is open


About a decade ago, an average Internet user might have heard of RSS. Really Simple Syndication, or Rich Site Summary – what the acronym stands for depends on who you ask – is a standard that Web sites and podcasts can use to provide their users with a stream of content easily understandable by many programs computer. Today, although RSS continues to feed many web applications, it has become, for most people, an obscure technology.

The story of how that happened is actually two stories. The first is a story about a broad vision of the future of the Web that has never succeeded. The second is a story about how a collaborative effort to improve a popular standard has been transformed into one of the most controversial areas in the history of open source software development.

In the late 1990s, between Netscape's IPOs and the Dot-com crash, everyone could realize that the Web was going to become an even bigger market than it was. already, even though they did not know exactly how. was going to get there. One theory was that the web was going to be revolutionized by syndication. The Web, originally designed to allow a simple transaction between two parties (a client looking for a document on a single host server), would be broken by new standards that could be used to repackage and redistribute entire Web sites through various channels . Kevin Werbach, who wrote for version 1.0, an influential investor newsletter in the 1990s, predicts that syndication "will become the central model of the Internet economy, allowing businesses and individuals to keep control of their personalities online while enjoying the benefits of scale and scope. "

The future was once so promising for RSS. What happened?

He urged his readers to imagine a future in which fencing aficionados, rather than go directly to an "online sports article site" or a "retailer". Fencing equipment ", could buy a new sword directly via ecommerce widgets integrated into their favorite website dedicated to fencing. . As in the world of television, where large networks subscribe to smaller local stations, web-based syndication would allow businesses and publications to reach consumers through a multitude of intermediate sites. As a corollary, this would mean that consumers would have significant control over where and how they interacted with a given company or publication on the Web.

The RSS was one of the standards that promised to deliver this subscribed future. For Werbach, RSS was "the prime example of a lean syndication protocol". Another contemporary article called RSS, the first protocol to exploit the potential of Extensible Markup Language (XML), a general-purpose markup language similar to the recently developed HTML. This was going to be a way for users and content aggregators to create their own custom channels from everything the Web had to offer. And yet, two decades later, after the rise of social media and Google's decision to close Google Reader, RSS appears as an endangered technology, used mostly by podcasters, programmers with technical blogs and occasional journalists . Of course, some people still rely on RSS readers, but stubbornly adding an RSS feed to your blog, even in 2019, is a political statement. This little mandarin bubble has become a nostalgic symbol of challenge to a centralized network increasingly controlled by a handful of companies, a network that hardly resembles Werbach's syndicated network of imagination.

The future was once so promising for RSS. What happened? Was its downfall inevitable or was it precipitated by internal conflicts that thwarted the development of a single SSR standard?

Muddy water

RSS has been invented twice. This meant that he had never had an obvious owner, a state of affairs that had engendered endless debates and acrimonies. But it also suggests that RSS was an important idea whose time had come.

In 1998, Netscape was struggling to build a future. Its flagship product, the Netscape Navigator web browser, which had already been preferred by more than 80% of Internet users, was rapidly losing ground against Microsoft's Internet Explorer. So Netscape decided to compete in a new arena. In May, Ben Hammersley, head of Ben Hamersley's Developing Feeds with RSS and Atom agency, assembled a team to start working on what was known internally. Two months later, Netscape announced the creation of its web portal, "My Netscape". would argue with other portals like Yahoo, MSN and Excite.

The following year, in March, Netscape announced an extension of the My Netscape portal called "Netscape Network". Netscape users could now customize their My Netscape page to contain "strings" with the latest titles of sites near the site. Web. As long as your favorite website has published a special file in a format dictated by Netscape, you can add this site to your My Netscape page, usually by clicking the "Add a channel" button that the participating websites were supposed to add to their website. interfaces. A small box containing a list of linked titles will then appear.

A My Netscape Network Channel

A My Netscape network channel for Mozilla.org, as it may seem interesting for users to add it to their My Netscape page.

The special file that participating websites had to publish was an RSS file. In the announcement of My Netscape Network, Netscape explained that RSS was the abbreviation for "RDF Site Summary." RDF, or Resource Description Framework, is essentially a grammar describing some arbitrary resource properties. (See my article on the Semantic Web if you find it exciting.) In 1999, the World Wide Web Consortium (W3C), the leading web standards body, was reviewing a draft specification for RDF. Although RSS is supposed to be RDF-based, the example RSS document published by Netscape did not use RDF tags. In a document accompanying the Netscape RSS specification, Dan Libby, one of the authors of the specification, explained that "in this version of MNN, Netscape has intentionally limited the complexity of the RSS format." The specification was version number 0.90, the idea being that later versions would allow RSS to align more closely with the W3C XML specification and the evolving draft of the RDF specification.

RSS was created by Libby and two other Netscape employees, Eckart Walther and Ramanathan Guha. According to an email that Guha sent me, he and Walther concocted RSS at the beginning with contributions from Libby; After the acquisition of Netscape by AOL in 1998, he and Walther left and that became Libby's responsibility. Guha had made a major contribution to the development of RDF and he and Walther wanted RSS to be an application of the format. But Libby later wrote that the initial vision for an RDF-based RSS had been reduced due to time constraints and the perception that RDF was "too complex" for the "average user". "

While Netscape was trying to win eyes in what we called "portal wars", elsewhere on the Web, a new phenomenon called "weblogging" was emerging. One of those pioneers was Dave Winer, CEO of a company called UserLand Software, which developed the first content management systems to make blogging accessible to people without much technical mastery. Winer ran his own blog, Scripting News, which is one of the oldest blogs on the Internet today. More than a year before the announcement of My Netscape Network by Netscape on December 15, 1997, Winer published an article announcing that the blog would now be available in XML and HTML format.

Dave Winer's XML format has become the Scripting News format. It was supposed to be similar to Microsoft's string definition format ("push technology" standard submitted to W3C in March 1997), but I could not find a file in the original format to verify this statement. Like Netscape's RSS feed, he structured the content of Winer's blog so that it could be understood by other software applications. When Netscape released RSS 0.90, Winer and UserLand Software started to support both formats. But Winer thought that Netscape's format was "terribly inadequate" and "missed most of the needs of writers and web readers." It could only represent a list of links, while the Scripting News format could represent a series of paragraphs, each containing one or more. connections.


Image: Shutterstock

In June 1999, two months after Netscape announced My Netscape Network, Winer released a new version of Scripting News, ScriptingNews 2.0b1. Winer claimed that he decided to go ahead with his own format only after trying, but without success, to make sure that nobody at Netscape cares about the defects of RSS 0.90. The new version of the Scripting News format added several elements to the element that brought the Scripting News format to parity with RSS. But the two formats have continued to differentiate in that the Scripting News format, which Winer has dubbed the "bold" syndication format, could include whole paragraphs and not just links.

Netscape posted RSS 0.91 the following month. The updated specification was a major topic. RSS is no longer synonymous with "RDF Site Summary"; he now represented "Rich Site Summary". All RDFs – and there were almost none anyway – were removed. Many Scripting News tags have been incorporated. In the text of the new specification, Libby explained:

RDF references have been removed. RSS was originally designed as a metadata format providing a summary of a website. Two things became clear: the first is that providers want more a syndication format than a metadata format. The structure of an RDF file is very accurate and must conform to the RDF data model to be valid. This is not easily understandable for humans and can make it difficult to create useful RDF files. The second is that few tools are available for generating, validating, and processing RDF files. For these reasons, we have opted for a standard XML approach.

Winer was extremely pleased with RSS 0.91, calling it "even better than I thought." UserLand Software has adopted it to replace the existing ScriptingNews 2.0b1 format. For a moment, it seemed that RSS had finally only one authoritative specification.

The big fork

A year later, the RSS 0.91 specification became terribly inadequate. RSS was trying all kinds of things that the specifications did not deal with. There were other parts of the specification that seemed unnecessarily burdensome – each RSS channel could only contain a maximum of 15 elements or links, for example.

At this point, RSS had been adopted by several other organizations. Other than Netscape, which seems to have lost interest after RSS 0.91, the main players were Dave Winer's UserLand Software; O'Reilly Net, who operated an RSS aggregator called Meerkat; and Overs.com, which also ran a news-based RSS aggregator. On the mailing list, representatives of these organizations and others have regularly discussed ways to improve RSS 0.91. But there were deep disagreements about the nature of these improvements.

At the root of this disagreement over namespaces, there was a deeper disagreement about the reason for being RSS.

The mailing list in which most of the discussions took place called the Syndication mailing list. An archive of the Syndication mailing list is always available. It's an incredible historical resource. It provides an instant account of how these deep disagreements eventually led to a political break in the RSS community.

Winer was on one side of the coming break. Winer was eager to evolve RSS, but he wanted to change it only relatively conservatively. In June 2000, he published his own RSS 0.91 specification on the UserLand website, which was meant to be a starting point for further development of RSS. It has not made any significant changes to the 0.91 specification published by Netscape. Winer claimed in a blog post accompanying his specification that it was only a "cleaning" documenting how RSS was used in nature, which was necessary because the Netscape specification was no longer maintained. In the same article, he asserted that RSS had succeeded so far because it was simple, and that some of the proposed changes in the Syndication mailing list would only make RSS "much more complex, and to IMHO, at the content provider level, it buys us almost nothing for added complexity. He was particularly sensitive to any plan that would add namespaces to RSS or reintroduce RDF formalisms that had been dropped before RSS 0.91 was published. (Namespaces would basically allow programmers to define RSS sub-formats, which means that new and exciting features could be added to RSS without everyone being in agreement on all the details. names would make the writing software more difficult to read.) In a message to syndication mailing list sent out about the same time, Winer suggested that these issues were important enough to cause him to create a fork :

I'm still thinking about how to move RSS feeds forward. I really want ICE content in RSS2, publish and subscribe is at the top of the list, but I will fight hard for simplicity. I like the optional items. I do not want to go through the namespaces and the road map, or try to make it a dialect of RDF. I understand that other people want to do that, so I guess we're going to have a fork. I have my own opinion on the orientation of the other fork, but I will keep them for myself at least for the moment.

Several other people, including Rael Dornfest of O'Reilly, Ian Davis (head of a research company named Calaba) and Aaron Swartz, 14, were fighting against Winer. It's the same Aaron Swartz who later co-founded Reddit and became famous for his hacktivism; in 2000, according to an email from Davis, his father often accompanied him to technology meetings. Dornfest, Davis, and Swartz all thought that RSS needed namespaces to accommodate the many things everyone wanted to do. On another mailing list hosted by O & # 39; Reilly, Davis proposed a namespace-based module system, stating that such a system would "make RSS as scalable as we'd like, instead of". Incorporate new features that would complicate the specification excessively. " camp "thought that RSS would soon be used for much more than syndication of blog posts, so namespaces, rather than complication, were the only way to prevent RSS from becoming unmanageable as it supported more and more cases of use.

At the root of this disagreement over namespaces, there was a deeper disagreement about the reason for being RSS. Winer had invented his Scripting News format to distribute articles written for his blog. Netscape had published RSS as "RDF Site Summary" because it was a way to recreate a miniature site in the My Netscape online portal. Some people felt that the original vision of Netscape should be honored. In writing on the Syndication mailing list, Davis explained that RSS was "originally designed as a way to create mini sitemaps" and that he now wanted, along with others, to develop RSS "to encompass more than information than just news headlines and to meet the needs of users ". for new uses of RSS that have emerged in the last 12 months. This overestimates the degree of unified vision of Netscape; Libby, in an e-mail to me, said that there was a dispute between the group "Let's Build the Semantic Web" and the group "Let's make it simple for authors" even during the development of RSS. In response to Davis' message, Winer argued for another original story: he claimed that his Scripting News format was actually the first RSS and that it was intended for a very different purpose. Since the people most involved in the development of RSS did not agree on who created RSS and why, choosing a range seemed inevitable.

The fork came after Dornfest announced a proposed specification for RSS 1.0 and formed the RSS-DEV working group (which would include Davis, Swartz and several others, but not Winer), in order to prepare it for publication. In the proposed specification, RSS again referred to "RDF Site Summary" because RDF was added to represent the metadata properties of certain RSS elements. The specifications recognized the name of Winer, attributing to him the merit of popularizing RSS through his "evangelization". But he also said that RSS technology could not be improved as Winer advocated. Simply adding more elements to RSS without providing scalability with a module system would "sacrifice scalability". The specification then defined a module system for RSS based on XML namespaces.

Winer felt that it was "unfair" that the RSS-DEV working group arrogated to itself the name "RSS 1.0". In another decentralization mailing list, he wrote that he had "recently stole a standard under a big name," which probably meant O'Reilly, who had convened the RSS-DEV working group. Other members of the Syndication mailing list also felt that the RSS-DEV working group should not have used the name "RSS" without the unanimous agreement of the community on how to advance the RSS. But the working group stuck with the name. Dan Brickley, another member of the RSS-DEV working group, defended this decision by stating that "the proposed RSS 1.0 format is firmly anchored in the original vision of the RSS, which itself had a long tradition going back to MCF (a precursor of RDF) and related. specs (CDF, ​​etc.). "He basically felt that the RSS 1.0 effort had a better claim to the RSS name than Winer, since RDF was initially part of RSS. The RSS-DEV working group published a final version of their specifications in December. In the same month, Winer released his own improved version of RSS 0.91, which he called RSS 0.92, on the UserLand website. RSS 0.92 has made several small optional enhancements to RSS, including the addition of the tag soon used by podcasters. RSS had officially poked.

The range might have been avoided if a better effort had been made to include Winer in the RSS-DEV working group. He obviously belonged there; He has been a major contributor to the Syndication mailing list and is responsible for much of RSS's popularity, as recognized by the members of the working group. But, as Davis wrote in an email to me, Winer "wanted to control and let RSS inherit his inheritance, so he was reluctant to work with us." Winer reportedly refused to participate. Tim O'Reilly, founder and CEO of O'Reilly, explained this in a UserLand discussion group in September 2000:

A group of people involved in RSS came together to start thinking about its future evolution. Dave was part of the group. When the group consensus turned in a direction that he did not like, Dave stopped participating and called it "O" Reilly's plot to take him the control of RSS, despite the fact that Rael Dornfest of O. Reilly was only one of about a dozen authors of the proposed RSS 1.0 specification, and that many of those who participated in his development had at least as much history with Dave as Dave.

Winer responded to Tim O'Reilly by writing the following:

I met Dale [Dougherty] two weeks before the announcement, and he did not say anything about the RSS 1.0 call. I spoke on the phone with Rael on the Friday before the announcement, but he has not yet said that they called him RSS 1.0. The first time I've heard about it, it's when it was announced publicly.

Let me ask you a direct question. If it turns out that the call to the new specification "RSS 1.0" has been planned privately, without warning or consultation, or so that the members of the syndication list have a chance to " 39 to agree or not, not just with me. go do?

UserLand has worked hard to create, popularize and support RSS. We went away and left the name to your guys. This is the upper level. If I want to do other work on web syndication, I have to use a different name. Why and how did it happen Tim?

I did not find a discussion in the Syndication mailing list on the use of the RSS 1.0 name before the RSS 1.0 proposal was posted. Winer, in a message to me, said that he was not trying to control RSS and just wanted to use it in his products.

RSS would be divided again in 2003, when several developers frustrated by the quarrels of the RSS community tried to create an entirely new format. These developers created Atom, a format that removed RDF but encompassed XML namespaces. Atom would eventually be specified by a standard submitted to the Internet Engineering Task Force, an organization charged with establishing and promoting the rules of the road on the Internet. After the introduction of Atom, there were three competing versions of RSS: RSS 0.92 from Winer (updated to RSS 2.0 in 2002 and renamed "Really Simple Syndication"), RSS 1.0 from the RSS-DEV and Atom working group. Today, we mainly use RSS 2.0 and Atom.


The proliferation of competing RSS specifications may have hindered RSS in a variety of ways that I will talk about soon. But that did not stop RSS from becoming extremely popular in the 2000s. In 2004, The Grouvy Today started offering its headlines in RSS and wrote an article explaining to the uninitiated what RSS was and how use. Google Reader, the RSS feed aggregator ultimately used by millions of people, was launched in 2005. In 2013, RSS seemed popular enough for the Grouvy Today, in its obituary for Aaron Swartz, to call technology ubiquitous . For a while, before a third on the planet had registered for Facebook, the RSS was simply the number of people who kept up to date on the news on the Internet.

Unfortunately, modern Web syndication is still done by one of the very few channels, which means that none of us "can retain control of their personality online" as Werbach imagined.

Swartz's obituary was published by the Grouvy Today in January 2013. At this point, however, RSS had taken a decisive turn and was about to become an obscure technology. Google Reader was stopped in July 2013, apparently because the number of users had dropped "over the years". This prompted several items from various outlets to declare that RSS was dead. But people said that RSS had been dead for years, even before Google Reader fired. Steve Gillmor, who wrote for TechCrunch in May 2009, said it was "time to completely abandon RSS feeds and switch to Twitter" because "RSS feeds do not cut it anymore." He pointed out that Twitter was basically a better RSS feed because it could show you what people thought of an article in addition to the article itself. It allowed you to follow people, not just channels. Gillmor told his readers that it was time to let RSS come back in the background. He finished his article with a verse of "Forever Young" by Bob Dylan.

Today, RSS is not dead. But nowhere is it as popular as it once was. Many people have explained why RSS has lost its general appeal. The most convincing explanation is perhaps that proposed by Gillmor in 2009. Social networks, like RSS, provide a feed containing all the latest news on the Internet. Social networks took over from RSS because they were simply better feeds. They also offer more benefits to the companies that own them. Some people, for example, have accused Google of closing Google Reader to encourage people to use Google+. Google might have been able to monetize Google+ to never monetize Google Reader. Marco Arment, the creator of Instapaper, wrote on his blog in 2013:

Google Reader is only the latest victim of the war unleashed by Facebook, seemingly accidentally: the battle to own everything. Technically, Google "owned" Reader and could take advantage of the huge amount of news and attention that circulated there, but that was at odds with their much larger strategy for Google+: they needed everyone to read and share everything via Google+ to compete with them. Facebook for advertising targeting data, advertising dollars, growth and relevance.

Users and technology companies understood that they used social networks better than RSS.

Another theory is that RSS has always been too geek for ordinary people. Even the Grouvy Today, which seems to want to adopt RSS and promote it to its audience, complained in 2006 that RSS is an acronym "not particularly user friendly" coined by "computer geeks". Before the RSS Icon Is Designed In 2004, websites like the Grouvy Today linked their RSS feeds with small orange boxes labeled "XML," which could only be intimidating . The wording was perfectly precise, however, because by clicking on the link, an unfortunate user was heading to a page containing XML. This super tweet captures the essence of this explanation of the disappearance of RSS:

Ordinary people have never felt comfortable with RSS; it was not really designed as a consumer technology and had too many obstacles; people jumped off the ship as soon as something better happened.

RSS might have been able to overcome some of these limitations if it had been developed further. Maybe RSS could have been extended in one way or another so that friends subscribing to the same channel could exchange their thoughts about an article. Maybe the browser support could have been improved. But while a company like Facebook was able to "move fast and break things," the RSS developer community was stuck trying to reach consensus. Lorsqu'ils n'ont pas réussi à se mettre d'accord sur une norme unique, les efforts qui auraient pu être déployés pour améliorer RSS ont été perdus au profit d'une duplication du travail déjà effectué. Davis m'a dit, par exemple, qu'Atom n'aurait pas été nécessaire si les membres de la liste de diffusion de Syndication avaient été capables de faire des compromis et de collaborer, et que «tout ce travail de nettoyage aurait pu être mis en place dans RSS pour le renforcer. Nous nous demandons pourquoi le RSS n'est plus populaire. Une bonne explication de premier ordre est que les réseaux sociaux l'ont supplanté. Si nous nous demandons pourquoi les réseaux sociaux ont réussi à le supplanter, alors la réponse est peut-être que les personnes qui s'efforcent de faire réussir le RSS rencontrent un problème beaucoup plus difficile que, par exemple, la construction de Facebook. Comme l'a écrit Dornfest sur la liste de diffusion Syndication à un moment donné, "actuellement, c'est la politique bien plus que la sérialisation qui est loin d'être simple."

Il nous reste donc aujourd'hui des silos d'informations centralisés. Malgré cela, le réseau syndiqué annoncé par Werbach en 1999 a été réalisé, mais pas comme il le pensait. Après tout, The Onion est une publication qui s'appuie sur la syndication via Facebook et Twitter de la même manière que Seinfeld s'est appuyée sur la syndication pour engranger des millions après la fin de sa publication initiale. J'ai demandé à Werbach ce qu'il en pensait et il était plus ou moins d'accord. Il m'a dit que RSS, à un niveau, était clairement un échec, car ce n'est pas maintenant "une technologie qui est vraiment le noyau de tout le monde du blogging, du contenu ou de l'assemblage d'éléments différents d'éléments dans des sites." Sur un autre plan, «toute la révolution des médias sociaux repose en partie sur la possibilité d’agréger des contenus et des ressources différents», d’une manière qui rappelle RSS et sa vision originale d’un réseau souscrit. Pour Werbach, «c’est l’héritage de RSS, même s’il n’est pas construit sur RSS».

Malheureusement, la syndication sur le Web moderne ne se fait toujours que par l’un des très rares canaux, ce qui signifie qu’aucun de nous «ne conserve le contrôle de sa personnalité en ligne» comme l’a imaginé Werbach. L’une des raisons qui explique cette situation est la rapacité des entreprises, parce que RSS, un format ouvert, ne permettait pas aux entreprises technologiques de contrôler les données et les yeux dont elles avaient besoin pour vendre des publicités. Elles ne l’ont donc pas soutenue. Mais la raison la plus banale est que les silos centralisés sont tout simplement plus faciles à concevoir que les normes communes. Le consensus est difficile à atteindre et prend du temps, mais sans consensus repoussé, les développeurs vont créer des normes concurrentes. La leçon à tirer ici est peut-être que si nous voulons un meilleur Web plus ouvert, nous devons mieux travailler ensemble.

Sinclair écrit sur l'histoire de l'informatique sur son blog, Two-Bit History. Suivez @TwoBitHistory sur Twitter pour ses derniers messages.