If you’re a regular reader of this blog, you probably know that attention is a recurring theme here. I’ve long been convinced that RSS aggregators can help people overcome information overload by first paying attention to what a user is reading, and then using that information to make better decisions about what might (or might not) be important to that user.
And I’ve also long been convinced that knowing what everyone is paying attention to, what your friends are paying to, etc., is a great way to uncover new and potentially important trends.
A big roadblock in this process has been the lack of a standardized format for storing attention data. While Attention.XML has been around for quite a while, for various reasons it hasn’t really caught on. As a result, many attention-related tools (including FeedDemon) rely on their own proprietary attention formats – which works fine for individual tools, but it doesn’t enable customers to easily share their attention data between services.
I looked into using APML as an attention format a while ago, but at the time I didn’t like the idea of storing a customer’s feeds in an OPML file, and then storing their attention data in a separate APML file. That’s easy enough for software to deal with, of course, but it would be a burden for less-geeky end users wishing to share their data. For that reason, I proposed an attention namespace for OPML – basically a way to store attention data within OPML. But that idea also never caught on for various reasons.
Fast forward to a couple of weeks ago, when I was prompted to take another look at APML by NewsGator Inbox’s Nick Harris. This led to a wider discussion about APML among NewsGator folks, including NetNewsWire’s Brent Simmons. We liked the "Really Simple Attention" approach proposed by the APML Workgroup, so we decided to tinker with APML despite my misgivings about storing subscriptions and attention separately.
The first step will be APML export, which is already underway. Once we’ve worked out a few details, we’ll also support APML import. This means you’ll be able to share your attention data not only between our tools, but also with any service that supports APML. Our hope is that by supporting APML, more third parties will be convinced to support it.
In addition, we’re working on ideas for supporting APML in NewsGator Online (not just for individual users who want to access their private attention data, but also for the service as a whole – ex: "what’s everyone paying attention to?"). The NewsGator Online piece is still in the baby stages so there’s nothing solid to announce yet, but we do want to let everyone know that it’s in the works.