Abusing XHTML
As small resumption to my previous post about XHTML/HTML here a small list of websites using XHTML that break when rendered on a browser in XHTML mode:
- adobe
- alistapart (although not everywhere)
- digg.com
- django webpage
- drupal webpage
- netscape portal
- php documentation
- python webpage
- ruby documentation
- sourceforge
- trac
Not that all my XHTML pages are valid, but if they fail… How should browser vendors implement XHTML if that would break the internets?
Ok, this shows only that these sites did not validate their stuff. So what - Anybody who delivers content with an xml content type will have to reckon with that.
It’s not that difficult to run the W3C log validator on one’s server each week and fix any broken pages. If xhtml content is not well-formed, then browsers *should* refuse to display it when it is delivered as application/html+xml. Otherwise xhtml support in mainstream browsers would become as spoiled as legacy html support.
If people *see* that their content is broken, because the browser brings this fact up, then it will be fixed. Otherwise bad practices will continue. But the sites that you mention would know what they are doing if they should start delivering their content with an xml-based mime type at some point.
Comment by faxi — Tuesday, October 30th, 2007 @ 11:38 pm