The RSS Blog

News and commentary from the RSS and OPML community.

With the release of Firefox 2.0 and IE 7.0, the importance of auto-discovery is moving from the realm of the uber-geeks to that of the mundane Web surfer. Without auto-discovery, IE 7.0 is gonna have a hard time finding your RSS feed to make it easy for users to subscribe to your feed. And you don't want that. You want to be the next Om Malik. That means, you have to do auto-discovery and you have to do it right. Let's start with the basics.

What is auto-discovery?

Auto-discovery is a technique that allows software to find your RSS feed from your blogsite address. Presented below is a fragment of the HTML you'll see behind The RSS Blog. You can check this out by using the View | Source menu option in Internet Explorer (if you use Firefox, then you likely already know this crap).

<html>
   <head>
      <title>The RSS Blog</title>
      <link rel="alternate" type="application/rss+xml" title="RSS"
         href="http://feeds.feedburner.com/TheRssBlog">
   </head>
   <body>

The element that allows for auto-discovery is the <link> element in the <head> section of the html source code of your website. This element should be present on the blogsite homepage, your blog entry webpages and any other webpages where you think a user might want to initiate a subscription to your RSS feed. You might be tempted to put the RSS feed of your blog comment feed or individual blog entry comment feeds. Trust me, this is a mistake. Your typical user is not gonna know the difference, in fact, I suspect I'm already confusing my own readers mentioning this crazy idea. Stick with your blog RSS feed. And while we're here, stick with one RSS feed. Don't provide both an RSS 2.0 and RSS 1.0 or Atom feed. One feed. Again, you'll only confuse the average subscriber.

Make certain the rel, type and title attributes are exactly as shown, including capitalizations. Change the value of the href attribute to indicate your preferred RSS 2.0 feed. Last, make certain the <link> element is inside the <head> element, that is, after <head> and before </head>.

OPML

Let me add that auto-discovery isn't limited to RSS feeds. You can also add an auto-discovery link to your blogroll or reading list. This is done in the same manner except that you change the media type in the link element. Don't worry about including both an RSS auto-discovery link and OPML in the same webpage, as I'm unaware of any problems that could arise. A sample OPML auto-disco link is shown below (which I stole from here).

<link rel="subscriptions" type="text/x-opml" title="Subscriptions" href="http://radio.weblogs.com/0001015/gems/mySubscriptions.opml">

I'm mentioning OPML auto-discovery only because I've noticed a bit of confusion in the blogosphere on how this should be done. In particular, note the type is "text/x-opml", not "text/xml+opml". The latter is invalid and doesn't make much sense at all.

To the uber-geeks reading this, please proof-read this for me. I want this to be ultra-accurate.

Update: Thanks to Marjolein Hoekstra for some great edits. Updates made in place.

Reader Comments Subscribe
Re: "Change the value of the href attribute to indicate your preferred RSS 2.0 feed."

I'd drop the 2.0 from this sentence, since you're advising people to pick one feed whether it's RSS 1.0, RSS 2.0 or Atom.

Rogers

Very good information on auto discovery. You are ultra-accurate.

Simon

 

Type "339":