A short while ago I added the “validate” link to the Atom 1.0 feed link so it was easier for me—and others—to check it, as I had forgotten to escape some ‘&’ in a URL a few days earlier. Normally Tidy takes care of the regular XHTML content and fix that sort of little errors that I may make, but the feed is just the raw data from the database with any errors that might be, and no correction.

So feedvalidator.org said everything was fine, yet rawdog, which Alex use, does not render entries from my feed as links. Strange.

Alex writes the author of rawdog and he finds the error, a misspelled MIME-type, ‘applicatio/xhtml+xml’, notapplication/xhtml+xml’ as it should be, for the XHTML chunks.

How did that ever get past feedvalidator?