XML sucks at (almost) everything it’s used for. Google has open-sourcedProtocol Buffer, a typed, backwards compatible, compact, binary data-interchange format. Combined with YAML for configuration and data-persistence files that need to be human readable, there’s even less reason to use XML for any data-serialization.
XML, in the form of (x)HTML, seems ok for markup, and the XML-based (x)HTML templating in Genshi is the best templating language I’ve ever used (and I’ve used XSLT, Mako, Mighty, PHP, and a few others). I wonder if the reason that HTML (and XML) templating is so difficult, and templating language code is often so ugly, is because XML is actually a poor solution for markup too. HTML is obviously here to stay, but it would be an interesting thought experiment to design a successor markup language that is not strictly hierarchical, more human-readable, and designed with templating in mind.
There are two quite different reasons for implementing HTML generation on a website. The first reason is to insert dynamic content, content that comes from a database or is algorithmically generated, into pages. The second reason is templating; to ensure that standard, site-wide parts of the HTML, such as headers and footers, are pulled from a single source. The goal of the first is to have a dynamic, database-driven site. The goal of the second is to avoid having to edit tens, or hundreds, of HTML files when the site design changes, and to avoid copy-and-paste coding.