In the 1800s, Italy was not the single nation as we know it today, but a splinter of several nation-states, some of which were under foreign control. A merchant marine captain and nationalist named Giuseppe Garibaldi would lead a movement that began at Sicily and moved north, eventually unifying the multiple city-states as a single republic.

Tim Bray, the co-author of the eXtensible Markup Language (XML), is trying to do the same for Real Simple Syndication (RSS) (define), except without guns.

Bray, director of Web technologies at Sun Microsystems, is co-chair of a working group for the Internet Engineering Task Force (IETF) (define) that's working to provide a single protocol for RSS.

Right now, there are nine incompatible RSS protocols from different development teams. And in some cases, the development teams don't even speak to each other. Two of the versions, RSS 1.1 and 3.0, were created as protests to how things developed on other versions of the protocol.

Despite this massive splintering, RSS has caught on as a popular means for sharing and syndicating data.

"RSS is clearly the most successful application of XML (define) in terms of the volume of data surging around, lives being touched and economic changes being made in publishing industries, but there's a lot more friction than there ought to be," said Bray.

Feeding on History

Although RSS has only caught fire in the last two years, it's been around, at least in concept, for a while.

When Microsoft released Internet Explorer 4.0 in 1997, it included a technology called the Channel Definition Format, which delivered content to the IE Active Channel feature in a similar method to RSS.

Also in 1997, Dave Winer, a programmer at UserLand Software, began designing his own XML syndication format for his blog.

In 1999, Netscape introduced RDF Site Summary 0.90, the first version of RSS, and the use of the acronym for the My Netscape portal. However, Netscape's days were numbered and it abandoned RSS. A working group would take up RSS while Winer worked on his own protocol.

RSS 0.90 would eventually become RSS 1.0, released in 2000, while Winer would release RSS 2.0 in 2002. In 2003, Winer declared RSS 2.0 finished and frozen and said all future development should be done on Atom, which was an IETF project. RSS may be frozen but it's still popular.

Today, RSS 2.0 and Atom are both in use by RSS readers and sites offering syndication, including internetnews.com.

"RSS 2.0 works fine, but everybody has agreed it should be stable and frozen and never changed again," said Bray. "Everybody has agreed if there's further work to be done it should be done under a new name and a new project, and that's what Atom is."

Atom "crystallizes what we've learned with RSS and unifies everything in one format, and secondly gives us a publishing protocol to empower anything to become a publishing engine and get more people online," said Bray.

Dealing with the RSS/Atom feed protocol is only half the battle. Bray said that an even bigger headache is publishing. All of the popular blogging sites, such as Blogger, TypePad, and LiveJournal, just to name a few, use a different protocol and they aren't interoperable.

"The authoring interfaces are terrible," said Bray. "We've gone from nowhere to the point where there's 6 million people who update their online presence once per week, but we're not going to get to 60 or 600 million unless we get the publishing interface a little better."

Bray praised Microsoft's move to put one-click blog publishing in Word 2007, which will come with Office 2007 early next year.

"That was absolutely the right thing to do," he said.

But the real killer app for blog publishing, he said, will be camera phones. People want to be able to take a photo with their camera phone and publish it to their blog with one click, he said, and the camera phone operators won't tolerate multiple protocols.

"You know they won't be implementing LiveJournal and TypePad and Blogger in millions of cell phones. They're going to pick one protocol. The choice facing LiveJournal and the like will be easy. Do you want all the people with cell phones to access [your site]? Then adopt Atom and you're done," said Bray.

Atom Ahead

At the moment, the IETF is close to finishing the Atom protocol and will begin interoperability testing. After that will come the Atom publishing protocol. Bray expects this to be done by the end of the year.

Then comes the really hard part: getting application vendors and middleware vendors in particular to adopt it.

Microsoft has gone crazy over RSS, making it an integral part of Windows Vista, Internet Explorer 7 and Office 2007.

Now comes the effort to convince other ISVs that Atom is an effective means for moving data around.

"There's a very strong movement afoot to use it for a wider range of tasks, like stock feeds. Anything with a series of events can use Atom," said Bray.

"Increasingly, [publishers] use XML but have to make their own wrappers. Atom has [wrappers] already, plus features like date stamp, a unique identifier, and digital signatures. So in a lot of cases, it will be a no-brainer to shift existing feeds over to Atom."

For Bray, evangelizing Atom feels like déjà vu.

"When we were getting ready to finish XML, where Atom is now, I was out on the road selling it. The kind of reaction we're getting to Atom now is very reminiscent to what it was for XML, where people are saying 'Oh yeah, I could use it for x,' where x is very different depending on who you are talking to," he said.