Julien Genestoux is the founder and CEO of Superfeedr. Superfeedr has provided a real-time infrastructure for RSS and Atom feeds since 2009. Julien is also a vocal open web evangelist and has been pushing forward the PubSubHubbub spec. Follow him on Twitter @julien51.
We all know that Google Reader, which used to be the most obvious RSS subscription tool, is now gone. At the same time, we see “follow” buttons on just about every website. We are at a tipping point with two contradictory trends: the decreased visibility of RSS feeds and the popularity of the “follow” feature. Google Reader and its market share was the biggest road block to a world where “following” can be both decentralized (open!) and ubiquitous.
RSS IS AWESOME BUT HARD
Even though it stands for “Really Simple Syndication,” RSS is a very complex tool to use for most. Don’t forget that many people have no idea what a browser is, so don’t expect them to know what a URL is, let alone a feed. Similarly, a vaguely orange icon with a radio wave on it is completely meaningless for anyone who’s not an advanced web user.
Let’s put ourselves in the shoes of a TechLone visitor who wants to follow the site in the new Digg Reader. First, they locate the orange icon on the right bar, the feed URL. They then click on that orange icon and arrive at a page that allows them to select what type of content they want to subscribe to. When they pick “TechLone,” they’re now shown a scary XML document. It’s hard to not hit the back button. They have to copy the address bar’s URL, head to Digg Reader, click on the add button, paste the URL and validate their new subscription.
To the advanced consumer of RSS feeds, this has become a long but relatively easy process, but for the majority without an understanding of the reason behind these steps, it is prohibitive.
The web needs a decoupled way to consume resources. We need to be able to consume feeds from TechCrunch, the New York Times or even my blog without needing an account or to agree on terms of service. RSS enables that, and in way that cannot be controlled or filtered by any given organization.
A lot of people rightfully argued that RSS complexity is what prevented its adoption by the masses and is one of the reasons Google eventually shut down its reader.
SUBSCRIPTIONS ARE UBIQUITOUS
In the last couple of years, we’ve seen a massive blooming of all kinds of “subscribe” or “follow” buttons. Twitter (one of the first promoters of the idea) allows you to follow other people. On Tumblr, you can follow other blogs. On Facebook, you can follow people and brands. On Quora, you can follow questions, and on GitHub, you can follow code. The list is endless. The “follow” feature has obvious benefits: engagement, notifications and, in the days of “infobesity,” it’s a great way to tailor one’s experience to their needs and interests.
It also helps most services build graphs to organize their data: on Twitter, it’s some kind of influence graph, and on Quora, it can be seen as a popularity graph, for example. Following is a vote of influence.
Unfortunately, these local graphs are not connected, which makes it hard to extract web-scale, meaningful information, which would help organize the world’s information. A well-known web-scale graph is the link graph, which was popularized and embraced by Google’s PageRank. By knowing which page links to which other page, Google was able to rank websites. Understanding what topics, pages and data sources are followed by whom can certainly also help categorize, rank and filter data.
RSS HAS THE SUBSCRIPTION GRAPH?
If we could combine RSS’s decentralized nature and the follow buttons that we’re seeing everywhere, then we can certainly hope to see another web-scale and meaningful graph in the form of the subscription graph. It would bring new types of discovery mechanisms. If two sources are followed by the same people, they could be grouped together even though they may never link to each other. There is also a strong prospective aspect to that, and we could detect trends faster by understanding what path the information takes from a follower to his followers. This signal is already exploited by Klout on Twitter’s data, but a web-scale version of that is also very promising.
Oddly enough we’re not too far from that. First, many services with a follow button do actually offer RSS feeds, as well. Tumblr, WordPress, GitHub, Quora and even Facebook pages have an RSS equivalent to their follow buttons. Most readers are still too “news-oriented,” but tools like IFTTT have interesting RSS-based recipes that can be used to create trigger actions when events happen, highlighting what events are worth being monitored by whom.
While adding RSS feeds to our “following” tool (I like that term more than “reader”) is still overly complex, solutions like SubToMe are here to simplify things without compromising the beauty of the decentralized RSS model.
Google Reader going away is clearly the end of the “RSS Reader” 1.0 era, and it opens the door for a much better, user-conscious era with easier tools to use and a more dynamic ecosystem. And if we stick to the RSS openness and avoid the numerous traps of silos, we can make the web even smarter by embracing a whole new set of linked data.