Semantic FaceWeb: RDfb is strong use of RDFa

Great news for Semantic Web, Facebook announced that they are going to use more structured metadata in shared web pages then just their original meta tags.

There are two parts to it – first they are going to provide storage about all things withing Facebook and on the internet in JSON format so people can write applications that use this data.

For example, public information about me on Facebook can be seen at this URL:
https://graph.facebook.com/sergey.chernyshev

And here how it looks:

{
   "id": "504972292",
   "name": "Sergey Chernyshev",
   "first_name": "Sergey",
   "last_name": "Chernyshev",
   "link": "http://www.facebook.com/sergey.chernyshev"
}

This way all you need to know is ID, which in my case is sergey.chernyshev or 504972292.

Another side of this is that Facebook is going to also store data from pages that were shared using their share button or added to interests or in any other way loaded into facebook’s system. They will be extracting this data using RDFa format and interpret it using a vocabulary that they call Open Graph Protocol (I will be calling it RDfb).

<html xmlns:og="http://opengraphprotocol.org/schema/">
<head>
<title>The Rock (1996)</title>
<meta property="og:title" content="The Rock" />
<meta property="og:type" content="movie" />
<meta property="og:url" content="http://www.imdb.com/title/tt0117500/" />
<meta property="og:image" content="http://ia.media-imdb.com/images/rock.jpg" />
...
</head>
...
</html>

This will allow sites to add a few lines of code to their pages to get them recognized as meaningful data instead of just human-readable documents – this will allow smart agents like search engines, mobile applications, embedded systems and just simple widgets to make sense of this data and help people see it in the context that is most usable for them.

I have a feeling that Facebook being a great driver for many business decisions related to promotion and marketing through social media, this change is going to be a best thing that happened to Semantic Web so far, might be even bigger then Yahoo’s and Google’s efforts in universal search.

P.S. Such a great present for my presentation about RDFa tonight ;)

2 thoughts on “Semantic FaceWeb: RDfb is strong use of RDFa”

  1. Sergei,

    In addition do not forget to highlight the fact that Facebook, is really demonstrating the power of structured data based on the Entity-Attribute-Value Graph Model. The commonality of the EAV model is what’s central to the ultimate Web of Linked Data bootstrap. The Model being distinct from associated Data Representations enables a variety of routes (Microformats, RDFa, OData, GData, various RDF formats) to the Web of Linked Data nirvana :-)

    The nirvana (in my eyes) is a point in time when all Data Objects (Entities) have unambiguous Names via Identifiers that resolved to Documents carrying their Structured Descriptions (in an EAV model Graph). The Representation of the these Entity Descriptions will be negotiable (not implying this to mean HTTP content negotiation, solely, but combination of auto-discovery patterns using and Link: response headers etc.).

    Links:

    1. http://bit.ly/cA0zxw — Data 3.0 Manifesto

  2. Absolutely agree. I think Facebook storing the data on their end for simplified development, even if not classic RDF approach, is still a very big move towards Semantic Web development – it’d strange if other big players haven’t provided this data in a similar fashion.

    And the one who thinks multi-dimensionally, e.g. vocabularies, inferencing, authentication, data versioning and so on, will win the game.

    BTW, Facebook is already providing authentication layer using OAuth.

Leave a Reply