RSS should be expanded to facilitate the rudiments of review data, given the extraordinary prevalence of the use of blogs for reviewing purposes. This could be easily facilitated with one new sub-element of item. 

<review>

Having the following attributes

• rating - numeric from 0 - 10, 0 being least. e.g. "10"
• category - from a common, pre-defined taxonomy of categories. e.g. "Book"
• guid - A guid for the item in question, formatted in a category specific manner.

All description and other links are facilitated by the existing RSS schema.

For example.

<rss>
  <channel>
     ...
    <item>
      <title>The Tipping Point: How Little Things Can Make a Big Difference</title>
      <description>Blah blah blah. Blah blah blah.</description>
      ...
      <review rating="7" category="book" guid="ISBN:0316346624" />
    </item>
    <item>
      <title>Casey's Food Pit</title>
      <description>Blah blah blah. Blah blah blah.</description>
      ...
      <review rating="2" category="restaurant" guid="905-555-5555" />
    </item>
  </channel>
</rss>

This is just off-the-cuff, as I was thinking of putting up some comments about the book The Tipping Point. Of course RSS took off because of its brutal simplicity, so I kept this as simple as imaginable (without resorting to namespaced schemas or complex structures to facilitate the varying data needs of each domain).

Surely there has to be widely accepted solutions in this space already?

My motivation is that I'm a big believer in owning your own data - Countless sites, including a lot of the Web 2.0 business plans, rely upon the fact that you will effectively hand over data to them, working as a free contributor to build their value. If you've reviewed books on Amazon.com, for example, or a digital camera on epinions.com, you've helped them build their own value. Sometimes this sort of arrangement is worthwhile (I'll happily host my pictures on Flickr because the return is worthwhile), but many times it is not.

Indeed I'd say that a good portion of the blog revolution was users taking control of their own data to some extent. Why post an insightful comment on some random message board, increasing the value to the message board, when there is little benefit for yourself? In the new era others can use your data - RSS aggregators for instance - but they have no monopoly or advantage outside of uniquely and innovatively providing some sort of added value.