Thursday, January 26, 2006

FeedFlare is big. Really big.

Forgive the multiple company posts, but our latest announcement is one of those deceptively simple things that takes a while to sink in. And I’m hoping that by documenting why I’m so excited about it that you might play around with it and use it (if you’re not already).

Last month, we announced FeedFlare. It generated quite a bit of excitement at the time, because it was the first fire-and-forget enhancement to your feed that let you dynamically add stuff into your feed — like exposing inbound links to the content via Technorati, one-click bookmarking to, and so on. Yesterday, the other shoe dropped, and we’ve now given you the ability to do the same thing back on your site.

Some of the techies who read this will say big deal, I hacked Movable Type/Word Press/whatever to do this ages ago. You’re right, no argument from me: if you know how to get in and hack your blog engine, FeedFlare doesn’t on the surface give you anything you can’t already do by yourself. However, the power of FeedFlare starts to get interesting when, in a couple weeks, we’ll open up the API. Now that’ll mean that with one click in your FeedBurner account, you’ll be able to dynamically add new web services — instantly — in your feed and on your website.

As Eric (our CTO) points out, there’s an additional benefit to our role in this process:

There’s one other very powerful thing about FeedFlare on the site: we are using all of the metadata and expressive power of the feed to inform the FeedFlare on the site. Since FeedBurner is processing your feed, all of the structured information is available to help generate the FeedFlare on the site. No having to scrape an HTML page like other scripts are forced to do — we can use the upstream information that simply isn’t available in the final, rendered page.

We’ve been brainstorming internally about ways to extend FeedFlare once the API opens up. We’ll publish that list (we’re gunning for 101 examples out of the gate) when we launch the API, but of course the community will figure out tons of ideas that haven’t occurred to us. I can’t wait.

On the tactical side, I’ve made some changes to my site to incorporate this. In no particular order:

  • Removed links to ‘trackback’ on the main blog page and individual entry pages. Now that FeedFlare dynamically updates the Technorati “flare” with all known inbound links, I don’t have to rely on trackback — which, since not all blog platforms support trackbacks, means that FeedFlare is now exposing more inbound links than Trackback ever did.

  • Added the ability to bookmark any page/item with, which should increase the likelihood that users will bookmark my content. This theoretically will expose more people to my content by making it easier to add tags to my pages.

  • Within the feed, added an ‘email this’ button that will let anyone reading an item in a feed reader forward the item to a friend.

When the API opens up, there will be a lot more options for enhancing a feed, and I’ll be making some changes to incorporate those new features. The good news is, once they are available, all I need to do is click a button at and my feed and blog will automatically inherit the new functionality — no rebuilds, no template hacking, nothing. Sweet.


  1. This sounds great! Anything that lets bloggers focus more time and energy on creating great content rather than dealing with the programming intricacies of their blog platform is a winner in my book.

  2. I agree, FeedFlare is huge! I cannot wait to see what additional interactivity features are created by developers since the API is opened. Also the updates you made to your blog are solid. ;)

  3. And I see you've removed those "this day in the history of tins" links from your feed too. (Yay!)

    One thing I wonder about with regard to FeedFlare or any other feed-enhancing widgets. How much of that do we want propagating through multiple feed aggregators. Not personal aggregators, but services that collect feeds and then redistribute them under a trusted banner. I'm part of the Corante network now. Does it make sense to add this extra stuff to the feeds that get distributed via that network? In fact, the Corante network is fed via FeedBurner. How should FB treat items that already have a FeedFlare in them? How should it treat items that already have similar "flare" added to them? For example, I think of tags listed at the end of feed content are "flare."

    I have a half-baked idea about wrapping the "flare" in additional metadata, but I see a big negative in that the less-ethical feed scrapers could more easily remove flare. For example, one of my regular feeds has a very clear byline added to the end of each post in order to more clearly indicate that he owns the content.

    I guess I'll have to follow this discussion on FeedBurner...

  4. Jack -

    Great feedback, thanks. Re: "On this day..." that had been in the feed, it was an idea I really liked, but whose execution was pretty poor. The idea was that as my feed circulation has grown (up over 700 readers now, from just 250 in 2004), "new" subscribers would potentially be interested in what I was writing about before. The reality was that though some people availed themselves of the links, they took up too much room and were repetitive. Look for something like this executed through FeedFlare after the API launches.

    As for your other questions, you raise several good points. Aggregated feeds need to be treated differently - and we'll need to have some logic that doesn't over-flare feeds through repetition. Having a way to exempt content that's already marked up makes sense, IMO.

    Yeah, some feed readers will block this stuff, but big deal. It's primarily a value-add for the person reading the content - if the reader renders that value-add invisible, they'll switch readers. Ultimately I think we'll see a lot of innovation on this concept, and it will be seen as something that the readers love as much as the publishers do.

    Keep in touch.