Perhaps I did something wrong, or failed to follow documentation rules, but the fact that the HTML files that I generated for Boost Exception were processed and wrapped in additional HTML caught me by surprise. Obviously I'm not against using uniform skin throughout Boost, but in general we can't just get HTML+CSS and wrap it with more HTML+CSS, there might be class name collisions, and whatnot. Besides the fact that now the Boost Exception documentation doesn't validate as XHTML 1.0, another problem is that some headings are simply function and class names, which the Boost CSS converts to allcaps, making them look like macros. It seems like other libraries have similar problems, I saw an instance of library documentation having its own tags indicating that it's HTML compliant, to which Boost is adding an additional XHTML 1.0 compliance tag, the end result being that the file is neither HTML or XHTML compliant. This is unfortunate and unprofessional. Is there a standardized XML format that can be used as documentation source? Emil Dotchevski Reverge Studios, Inc. http://www.revergestudios.com/reblog/index.php?n=ReCode