When Google announced Rich Snippets just over one year ago on May 13 2009, they lit a fuse. A year later we saw the explosion when Facebook and Twitter both made moves to support Semantic Web standards. Who could have forecast that?
We are now seeing the great convergence. The new web looks a lot like a triumvirate - Google, Facebook and Twitter. There are other huge companies doing great stuff, but this triumvirate moves the market and defines standards adoption. All three do social and all three do search. So they all both publish and consume content.
On April 21st, we wrote "Facebook just nailed the semantic web". (Actually, we quoted Alex Iskold's tweet on that subject and then tried to figure out and explain why he was right).
Since then, we have seen momentum around Twitter Annotations and Google is now talking about how much traction they are getting with Rich Snippets.
In this post we try to dig below the covers to see how this might play out in the emerging new battle for leadership of the semantic search engine market.
RDFa: What And Why
The triumvirate are agnostic on metadata standards. Their approach is "we support whatever works" - RDFa, Microformats, OpenGraphProtocol, AB Meta, Activity Strea.ms, whatever. Google adopted that agnostic approach when they launched Rich Snippets. That is the right strategy for a search engine. You search what exists, not what somebody says should exist.
But that may not be the right strategy to win the semantic search engine battle.
First, if you are new to RDFa, here is a good intro:
Search engines need to be standard agnostic. But publishers and webmasters need a standard. They need to decide what metadata to publish. This is an action that will be taken by millions of people every day, so it has to be obvious and simple, low friction.
This is what Facebook understood and why they may emerge as the winner in the semantic search engine game.
And the fact that Facebook is moving closer to RDFa will impact webmaster/developer behavior and that will influence the moves of Google and Twitter.
Facebook Open Graph Protocol
Nick O'Neil at AllFacebook likes to say "Like is Replacing Link". That resonates. Like is lower friction. It is also easy to rank. It is obvious what you mean. (Yes, we also want Dislike and some of us who "do nuance" would like that, but it is the simplicity of Like that gets the traction).
But you have to Like something. It has to be clear what you mean, or Facebook will simply be aggregating junk data. That is where a standard such as RDFa comes in. That is why it is so cool that Facebook is listening to the webmaster/developer community around Drupal when they advocate for RDFa (as we reported here yesterday).
Google Rich Snippets
Here is Kavi Goel introducing Rich Snippets over a year ago:
To nobody's surprise, it has taken off like a rocket. The use cases are mostly reviews. Here is an example of how you can put rich snippets into WordPress without editing your theme. This is simple, accessible to any blogger/webmaster, totally mainstream:
But now that Facebook is supporting RDFa and Twitter is "sort-of supporting RDFa" (see below), the amount of RDFa on the web will grow and so Google will support more RDFa.
Support in this case, means publishing content on the Google search page. That is a controversial move. They have to tread a fine line around "fair use" of content. Oh, it is tough being a gorilla, when all your moves get so carefully analyzed by regulators
In other words, Google has to figure our what to do with RDFa data. That will almost certainly be on a case by case basis (what works for restaurant reviews might not work for healthcare data and vice versa).
But being a foolhardy fellow, I will make this forecast:
"during 2010, Google will announce more support for RDFa in Rich Snippets".
And, during 2011, a start-up will leverage that to a huge success.
It is easy to forget that Twitter is actually quite a small company and they have to compete on a lot of fronts. This week the top priority has probably been simply keeping the service up and running during the World Cup traffic onslaught. And they have to make search an awful lot better; search is the key to monetization and now that Twitter have started on monetization they need to scale it fast or lose credibility.
So if you look at the A List of the Twitter management team, Twitter Annotations is probably not top of the list. The strategy seems to be to let the market create solutions. And the market will create solutions as this presentation by the TwitLogic guy shows:
In simple terms, TwitLogic turns "twitter streams to RDF streams". So Twitter does not have to worry about being RDF compliant, the market will fill in for them. At some point, this will become a priority for Twitter and then they will either a) buy TwitLogic, b) partner with them or c) copy and bury them. They need to do more of a) and b) so that the developer community remains motivated to fill in their holes.
There will be more of this "real time semantic web middleware".
This open market approach may be a winner for Twitter. Time will tell. It probably their only strategy as Google and Facebook are so much bigger.
And The Winner Is.....
Ta da, Facebook. At least today. This game is like Basketball, fast-moving and superb athletes. So expect some changes.
Why Facebook? Three reasons:
1. They have the scale.
2. They have a clear strategy, using Like to replace Link, making Like the recommendation engine of the web (as recommendation engine is the key to ecommerce and that is where the money is).
3. They understand that they need to use RDFa as a competitive advantage against Google.
But the other players have some great moves:
• Google: They have the SEO mainstream, as shown by that WordPress plug-in. That is where Drupal is making the cool moves. Their support for RDFa in Drupal 7 will enable webmaster to get lots of traffic from Facebook and that will make them add more RDFa and then Google will take notice...
• Twitter: They are trusting to the "magic of the ecosystem" to fill in the holes. Magic is of course rather unpredictable
Yahoo Search Monkey: Honorable Mention Or Dark Horse?
I kicked off my post with:
"When Google announced Rich Snippets just over one year ago on May 13 2009, they lit a fuse."
That of course is not true. Yahoo Search Monkey was the pioneer. Today that looks like a footnote in history (like Xerox Parc in the Apple and Windows story). But Yahoo retains a loyal following among developers. They could be the neutral Switzerland in the semantic search engine wars, offering a search infrastructure for sites/developers to make money from (which they share in some way). Count me in the contrarian crowd looking for a Yahoo resurgence...
The Big Winners: RDFa And All Of Us
When gorillas compete, everybody else wins. The logic of the market is increasing support for RDFa by Google, Facebook, Twitter and therefore everybody else.
That is a win for open standards and that is a win for all of us, who can publish RDFa and search RDFa and build tools that make publishing and searching RDFa easier.