Update: I just discovered that Tantek Cilik has incorporated a rating sub element into his hReview microformat spec. It doesn't however, solve the common variation problem on ratings which is that most modern ratings tools prefer the thumbs up-no-opinion-thumbs-down approach. Maybe he'll add that at a later stage. Here's some sample hReview code:
Excuse the bizarre title, but it'll make sense in a minute I promise. So today I was meeting with Chris Jackson of MetaBroadcast and George Wright of BBC RAD. We were discussing a few projects in the pipeline, when a thought struck me. Is there a web standard for displaying and accepting ratings data? Would this benefit the Web? The idea struck me from talking with Chris and George about MetaBroadcast's URIPlay project, which aims to standardise meta data around programmes and their rights policies to facilitate the discovery of content in the brave new semantic world. One of the things that I would like to see implemented on the channel4.com site is a wdget which allows a user to beam up some preference information about the programmes that they have enjoyed watching, or share with a selection of their friends. That's nothing new, of course - Last.fm has been doing i for a couple years. However, I don't really see the value of the ratings (or 'favouriting') if the results of those ratings are constrained to channel4.com. As people using the widget could only get the benefit of their sharing their preferences on c4.com, few people would use it, in my estimation. So the obvious development of the concept is a cross-broadcaster service that allows the user to rate or hate content on all the big content publishers. That would be worth using. So anyway, that's a project that's been on my mind a lot lately. But it got me thinking that if someone were to produce a shared recommendation API for broadcasters to use, what would stop them for producing a similar API to rate anything, and store all those ratings in a database somewhere for the benefit of the general public? What if Google were to release a Ratings API that standardised how developers implemented popularity-tracking services? But of course, simply applying a standard set of microformats could go someway towards semantically addressing the problem. If whenever I displayed a rating of something on my site, say, I used some standardised markup like:
Then any crawler looking for those standard attributes in the code could grab some valuable data from my site, and store it, analyse it against similar data, and start to build up a nice store of useful information about the popularity of content across the web. Going to do a bit of digging on this and see if anyone's done something similar already...
<div class="hreview"> <span class="reviewer vcard"> <span class="fn">anonymous</span>, April 18th, 2005Ying Xiong (HERO) </div> <div>Rating: <span class="rating">4</span> out of 5</div> <div class="description"><p> This movie has great music and visuals. </p></div> </div>
Excuse the bizarre title, but it'll make sense in a minute I promise. So today I was meeting with Chris Jackson of MetaBroadcast and George Wright of BBC RAD. We were discussing a few projects in the pipeline, when a thought struck me. Is there a web standard for displaying and accepting ratings data? Would this benefit the Web? The idea struck me from talking with Chris and George about MetaBroadcast's URIPlay project, which aims to standardise meta data around programmes and their rights policies to facilitate the discovery of content in the brave new semantic world. One of the things that I would like to see implemented on the channel4.com site is a wdget which allows a user to beam up some preference information about the programmes that they have enjoyed watching, or share with a selection of their friends. That's nothing new, of course - Last.fm has been doing i for a couple years. However, I don't really see the value of the ratings (or 'favouriting') if the results of those ratings are constrained to channel4.com. As people using the widget could only get the benefit of their sharing their preferences on c4.com, few people would use it, in my estimation. So the obvious development of the concept is a cross-broadcaster service that allows the user to rate or hate content on all the big content publishers. That would be worth using. So anyway, that's a project that's been on my mind a lot lately. But it got me thinking that if someone were to produce a shared recommendation API for broadcasters to use, what would stop them for producing a similar API to rate anything, and store all those ratings in a database somewhere for the benefit of the general public? What if Google were to release a Ratings API that standardised how developers implemented popularity-tracking services? But of course, simply applying a standard set of microformats could go someway towards semantically addressing the problem. If whenever I displayed a rating of something on my site, say, I used some standardised markup like:
<div class="rating"> <span class="rt_object">Video Title</span> <span class="rt_obj_url" title="http://www.somevideourl.com" ></span> <span class="rt_type_out_of_five" title="5">5</span> <span class="dt_rating" title="2009-16-01">January 16</span> <span class="rt_user" title="username">username</span> </div>
