Today I came accross an interesting interview to Amit Singhal from Google, lead developer of Google’s real-time search. He explained how a “tweet” is ranked by Google and why a uses has a higher reputation than another user.
Amit’s analysis is pretty clear: the approach used by Google is very similar to the one used for PageRank, the popular algorithm Google uses to rank websites.
In the case of tweets, the key is to identify “reputed followers,” says Amit Singhal, a Google Fellow, who led development of real-time search. (Twitterers “follow” the comments of other Twitterers they’ve selected, and are themselves “followed.”). ”You earn reputation, and then you give reputation. If lots of people follow you, and then you follow someone–then even though this [new person] does not have lots of followers,”
While it makes sense, I disagree on the overall approach. If I put a link on my website/blog post/online article, I’m explicitely “endorsing” the content at that URL. This way, if that website is linked a lot of times by many different other websites, its pagerank will be affected positively.
On the contrary I follow a Twitter user, I’m not explicitly endorsing him/her, neither necessarily paying attention to what he/she tweets. At the same time, If I follow a certain user and I have a lot of followers, this doesn’t mean that if “my reputation” is very high I should give “more” reputation to the user I’m following.
All that being said, when we designed my company’s realtime Twitter search engine Tweefind, we chose a more complex but accurate way to assign a rank to a Twitter user, that takes into account almost 30 parameters and followers/following contributes just a bit to the total rank. This is because we give more value to how Twitter users engage and participate to conversations, how much their tweets get retweeted and how much they get involved in the Twitter experience.
To say it in other words, I don’t think my rank should be higher if I get followed by Techcrunch, Mashable, Alyssa Milano or Robert Scoble, but it should be affected just in case those (for example) users interact with my twitter stream, favorite my tweets, retweet my content and so on. That’s what matters on Twitter and I’m honestly a bit concerned that Google is not taking all this stuff into account. This post by Google Operating System somehow confirms this approach.
What do you think? Is Google not telling everything about how they rank tweets? Let’s start a discussion, leave a comment down here, I’m very interested to know your thoughts on this subject.
Related articles by Zemanta
- Real Time Search, yet another oxymoron? (ecairn.com)
- Google Real-Time Irrelevance (googlesystem.blogspot.com)
Why I run my blog on Thesis Wordpress Theme
Thesis gives my blog a very professional look with very clean, easy-to-read layout and SEO friendly design. Take a look at my post explaining why I chose Thesis for my blog. Get your Thesis Theme today!
Why I host my blog on Media Temple
Media Temple Grid-Service is a hosting solution which guarantees speed and reliability to my blog. Unlike other shared hosting plans, Mediatemple Grid-Service eliminates the “bad neighbor effect” by distributing the load over several servers instead of confining it to one machine. This redundancy means more stability and virtually no downtime. Get Media Temple Hosting services now!



![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=39b7cef1-cb6d-41d5-b119-32b41fa47a34)

