Feedburner CEO Dick Costolo has just posted what I think is a milestone post for RSS and Web 2.0: How feeds will change the way content is distributed, valued, and consumed. In it he expounds on the future direction of his company Feedburner, which I've long considered to be the leading company in the RSS Publishing space. (NB: this post is NOT about the other RSS news of the day, Microsoft's Simple Sharing Extensions. I'll address that in a separate post).
There are 3 themes in particular from Dick's post which grabbed my attention:
1) RSS feeds encompass much more than blogs now
2) Feed becomes input to content on the site
3) Focus on the feed item - attach threads to the item and track it across the Web while it gets remixed and re-published.
With regard to 1), many of us have been noticing the trend for RSS feeds to extend into non-blog content. 2005 has really been the year in which that trend has solidified, as Dick's venn diagram perfectly illustrates:
Feed becomes input to content on the site
This is where it gets really interesting. Feeds becoming input to the website reminded me of Erik Benson and others' experiments with the Bloglines API back in February 2005. At that time, Erik re-modelled his whole blog so its content was entirely made up of inputs from his many feeds. He did that because his content was being published all over the Web - in del.icio.us, Flickr, 43Things, LiveJournal, other blogs. So feeds became his method for keeping track of all his content and bringing it all together.
It makes total sense for Feedburner to be right in the middle of this 'pulling together' of feeds from all over the place - and other feed input mechanisms. People like Erik have the programming knowledge to manage their feed 'inputs' this way, but the vast majority of people (like me) would rather a service like Feedburner did that work for us - via our Feedburner feed and some simple hook-ins to systems like Movable Type, Wordpress and Typepad.
"The explosion I am talking about is the shifting of a website's content from internal to external. Instead of a website being a "place" where data "is" and other sites "point" to, a website will be a source of data that is in many external databases, including Google. Why "go" to a website when all of its content has already been absorbed and remixed into the collective datastream."
His post specifically referenced Google, but I think this trend is much larger than even Google. The thing which is going to tie all this together is of course feeds. Mainly RSS, but perhaps Atom's much-vaunted extensibility will come into play too.
Focus on the feed item
This gets to the heart of the matter and I think Feedburner is onto something big here. Feedburner now views the item (e.g. a single post from your blog, or a specific search result in a topic feed) as "the atomic unit of measure in the feed", which will in turn lead to Feedburner managing syndicated content "at a more atomic level by attaching 'threads' to the item." It reminded me of the Design for Data and "content will be more important than its container" themes I was big on at the end of last year and beginning of this (and which I will be re-focusing on now). Incidentally those R/WW posts from a year ago led to a collaboration with Joshua Porter on a Digital Web article, which led to an O'Reilly Media book contract. But I digress...
If you think about it, focusing on the feed item is a profound change in how we think about RSS feeds. Up till this year, most of us thought of RSS feeds as a way to subscribe to single sources of content. But over 2005 it's become apparent that content is being remixed, mashed up and re-published across many sources - leading to heated ethical debates over content rights and confusion amongst publishers on how to 'monetize' (sorry I can't help but use that word) their content. Fred Wilson had a nice post on this theme recently, entitled The Future of Media (aka Please Take My RSS Feed).
The 2.0 Toolset
Dan Saffer recently explored this issue from a different angle, a post entitled The Web 2.0 Experience Continuum. Dan's post is all about the need for a next-generation tool set to deal with what he calls semi-structured and unstructured Web experiences. He wrote:
"The tools we’ll use to find, read, filter, use, mix, remix, and connect us to the Internet will have to be smarter and do a lot more work than the ones we have now."
What he's referring to is at the aggregation and filtering level, whereas Feedburner is at the feed/item level. So I think there are opportunities for developers to create those Aggregation 2.0 tools, which will complement what Feedburner is in the middle of building.
I've got my own ideas on what the next generation of Aggregator/Filter tools will look like, so perhaps I'll even get stuck in and try to build part of this vision out myself. Exciting times!