One feature that helps add weight (generally) to a LinkedIn Profile is ‘Recommendations’. I’m not getting into debates on how it’s used etc, that’s a subjective thing, but someone else acknowledging that the concerned person has certain skills does help. Facebook recommends friends, Twitter recommends users to follow. These are three layers – in LinkedIn its a human, in Facebook its an algorithm basis the user’s location, friends etc, and as for Twitter, well, Twitter just decides – no algorithms. But its ok, we recommend links to each other on Twitter. 🙂
A few activities recently made me think of recommendations. Two from Google and one from Facebook. A TechCrunch article from a few days back states that Google Friend Connect now has a widget that can help publishers know (and display) which parts of their websites their visitors like best. So it helps both parties. I’m guessing it should also help Google figure out a little more data on who reads what where, and therefore some thing that can be used to improve Ad Sense’s effectiveness. 🙂
One of Google’s services that uses a recommendation mechanism is Google Reader. Google has now added a feature on Reader that lets you know which of your friends are still worth following on Reader, basis your consumption of their shares. I wonder if they’ll utilise this data for new users – eg. if A and B are existing users and C joins the service, will Google use the A’s and B’s data to help C start off? I also think users should have the option of sharing their own trends data with each other, tools can be used to enhance utility – eg. if i know that 90% of my friends are following TC, then I might share less of TC items.
Meanwhile, RWW thinks that Facebook has to be working on some recommendation technology. With those thumbs up and down signs on ads, I won’t be surprised if Facebook uses that on friends – ‘Manu liked this ad’ (so we’re serving this to you, since you’re his friend) and one more ‘rebellion’.
Also, from RWW, a related topic, for a larger perspective – Linked Data. “Sir Tim Berners-Lee, the inventor of the Web, gave a must-view talk at the TED Conference earlier this year, evangelizing Linked Data. He said that Linked Data was a sea change akin to the invention of the WWW itself.” We are moving towards a web that’s increasingly inter connected.
That made me think – we’ve reached a state where you can now login to Facebook with your GMail id (not vice versa yet), thanks to its working with OpenID. There are tools on existing social networks (and new services) for location based social networking. Made me think of the potential of a larger recommendation based web experience, that can then spill over on to real life. Recommendations are already being used, even in online commerce.
But what it actually made me think is about a larger system where say, Facebook, the ad publisher and I will all share revenue if the friend does some positive action on the ad served to him, thanks to me. And of course, Google will then use this info to serve ads to me later, or utilise this on its own Friend Connect + iGoogle+ AdSense . 😉
Virtually connected lifestreams and real money. The friends of friends of friends connection utilised upto a huge degree (with privacy controls) – its not a real social connection, only an algorithm that would calculate relevance basis the degree of separation and the history of activities. Recommendations of ideas, links, ads, people, jobs, music, books and any kind of products, services etc.. an algorithm boost to ‘serendipity’, if you will 🙂 It even works the other way, so if you say, log in to a site to check out products, it immediately searches to see if there’s a recommendation it can push at you. Trust automatically plays a key role, and how well past recommendations have worked for you.
Meanwhile, let’s hope that Google doesn’t make a social algorithm to top the one they’re working on now – to identify which of its employees are likely to quit. A recommendation feature that allows one employee to suggest another would be a Google killer. 😉
until next time, ahem, some social advertising -I’d recommend watching this space – for a virtual interview 😉
One thought on “What do you recommend?”