Polluting markup is not the solution. It looks really ugly and makes maintenance a nightmare. Separate markup from data (and style, scripts, etc) for a cleaner app.<p>I've always been a proponent of placing all data in a script using json which you can consume easily without screen scraping.<p><a href="http://mylittlehacks.appspot.com/dataislands" rel="nofollow">http://mylittlehacks.appspot.com/dataislands</a>
why bother? because google, for example, will use some of the annotations:<p><a href="http://www.google.com/support/webmasters/bin/answer.py?answer=99170" rel="nofollow">http://www.google.com/support/webmasters/bin/answer.py?answe...</a>
Recent discussion here about microdata: <a href="http://news.ycombinator.com/item?id=1673623" rel="nofollow">http://news.ycombinator.com/item?id=1673623</a>
I have always liked (and often written about) an alternative: for HTML: <a href="http://example.com/test.html" rel="nofollow">http://example.com/test.html</a>, for semantic data: <a href="http://example.com/test.n3" rel="nofollow">http://example.com/test.n3</a> (N3 RDF format).<p>If a web app creates dynamic data at URL: <a href="http://example.com/stockfeeds?symbol=APPL" rel="nofollow">http://example.com/stockfeeds?symbol=APPL</a> then perhaps use something like: <a href="http://example.com/stockfeeds.n3?symbol=APPL" rel="nofollow">http://example.com/stockfeeds.n3?symbol=APPL</a> to get RDF. Also, have standard (or commonly used) meta data to point to an equivalent RDF URI in generated HTML.