Preparing your feeds for IE7

Internet Explorer 7 logo Internet Explorer will be released in just a few weeks, pushed to Windows XP users as a critical update. The Windows RSS Platform ships as part of IE7 and will likely become the most popular desktop aggregator by the end of the year. Are you ready for the switchover?

There are changes to CSS and JavaScript handling and an OpenSearch search box you should probably code against if you would like quick and easy access to your site and its archive. I’m mainly interested in the changes in feed syndication so I’ll walk through some areas that might trip you up as a publisher.

IE7 feed view

Valid XML only

Is your feed valid XML? If you or your customers are outputting content with invalid characters, an undefined namespace, or a non-breaking space ( ) the Windows RSS Platform will disregard your feed updates. A snapshot of Google Reader’s subscriptions last December found about 7% of the feeds it indexes are not well-formed XML.

Use modern feed formats

The platform includes support for feed formats RSS 2.0, RSS 1.0, Atom 1.0. If you are still outputting in RSS 0.91, RSS 0.92, or Atom 0.3 IE7 will still support the format, but you are encouraged to upgrade to a more recent feed format for the best support. Feeds that reference a DTD are considered a potential security issue and the feed parser will reject the feed and display an error message.


Feed button IE7

Can web browsers easily find your feeds? Internet Explorer 7 tries to auto-discover feeds referenced as a link alternate in your HTML. IE7 mimics Firefox’s auto-discovery behavior, so if you notice your feed(s) lighting chicklets in Firefox you should be all set. Your web server can help identify feeds by serving the correct MIME types for each feed type such as application/atom+xml or text/xml. Browsers take a number of steps when trying to identify your feed. If you produce better output the browser does less work!

Check for valid feed names

Do you have a valid feed name? The Windows RSS Platform supports feed names between 1 and 120 characters in length and may not contain a back-slash (“”) or Unicode control characters in range 0-31.

Check for valid feed markup

The Feed Validator can help you find more issues in your feeds that might cause problems for feed parsers. The feed validator project is open-source and you can run your own local copy using Python.

  • Posted
  • Updated at
  • Comments [3]


Commentary on "Preparing your feeds for IE7":

  1. Thomas Hoey on wrote:

    Have used IE7 Beta for a while. Went back to 6.0 recently. 7 (beta) is VERY slow and is still bothered by phishing and spywre problems. When Firefox has almost eliminated all problems, it boggles the mind that Microsoft cannot do better.
    BTW. Vista REALLY sucks. It can’t get much worse.

  2. paul on wrote:

    Probably okay for consumers, but does the business world really want RSS aggregation in the browser? Wouldn’t it make more sense to focus on aggregation in Outlook?

  3. Matt Terenzio on wrote:

    Thanks Niall. Great post.
    I find the OpenSearch very interesting.
    One thing about IE7 beta that I thought was a mistake was the fact that you can’t apply an XSLT to an RSS page since IE recognizes it and uses it’s internal display.
    Well, I think they should look for a style sheet first, and if one is not present, then use their own rendering.
    Browser side XSLT is one of the coolest and under-utilized cross-browser tool around and I’d hate to see this behavior move over to other formats like OPML.