When I click on the RSS link, my web browser tries to download it instead of display it.


Textpattern identifies its RSS and Atom feeds as “application/rss+xml” and “application/atom+xml” respectively. This is so that applications can distinguish between the different types of feeds without having to guess [1].

Different browsers handle “application/xml” in different ways. Mozilla and Firefox prompt the user with a file download dialog box.

The important thing to remember is that XML feeds are not web pages. They are not indended for viewing in a web browser.

[1]. The “old-fashioned” method was to identify feeds as “text/xml”. This merely identifies the document as XML, not as a news feed. Nor does it provide any way to distinguish between RSS and Atom, and can lead to character set encoding problems.