Richard (on having the img on a separate dd): There is no universal solution to it. It depends on content. So, if you are tagging a book, with its synopsis and cover:
<dl> <dt>Book Title</dt> <dd><img></dd> <dd>Description</dd> </dl> <dl> <dt>Book Title</dt> <dt>ISBN</dt> <dd>Description</dd> <dd><img></dd> </dl> <dl> <dt>Book Title</dt> <dd> <img> <p>Description</p> <p>Description</p> </dd> </dl> To me all are good semantic options (remember this is just an example for a books). The cover on a single dd? I'm not sure, after all the you can't describe a book by the cover (ha, ha, I'm so funny) To me the cover would be part of the description along with the synopsis. Or not, but I do not see it neccessary to have every descriptive item in its own dd. And look at the example I provided with the ISBN on another DT; after all you are describing an item equally identifiable by both ISBN and Title, but you could say: "no, the ISBN is not that important in this specific page I'm coding" then do not put it inside a dt, semantics are not context-independent. The author of the book probably deserves its own dd, but then again, the author, synopsis, publisher are just part of what describes a book. What do you think? I'm having now a similar problem on a magazine: ______ | | ARTICLE TITLE (variable length) | photo | variable length short desc |_____ | ______ | | ARTICLE TITLE (variable length) | photo | variable length short desc |_____ | When you have this scenario. with more than 1 set of article, desc, img, css get tricky if the sets are on the same dl. So I'm considering this: (BTW Why can't we wrap dt/dd with divs? Divs have no semantics, they should be allowed everywhere; I gues that's why they've come up with <di> in XHTML 2.0) <ul> <li> <dfn>ARTICLE TITLE (variable length)</dfn> <img> <p>variable length short desc</p> <p>variable length short desc</p> </li> <li> <dfn>ARTICLE TITLE (variable length)</dfn> <img> <p>variable length short desc</p> </li> </ul> Is it pushing too far? Oh! I hate applying semantics on non semantic languages. ****************************************************** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list & getting help ******************************************************
