It’s been a long time coming, but Digg is finally rolling out its “Recommendation Engine” to registered users this week. The basic idea is that the service will look at what stories you have dugg (the term for when you give an ‘up’ vote to an item) and based upon that, find similar users to you, using them to send you new stories.
Flat out: This is the feature that can make Digg useful to me again.
The premise and promise of Digg has always been great: Utilize the “wisdom of the crowds” to find the best and most interesting news. Certainly this will bring about different items than any one editor or even a group of editors for a publication or website could. The problem in the past several months is that Digg has gotten too big for its own good.
With over 16,000 stories now submitted to Digg each day (interestingly in the video embedded below, Digg co-founder Kevin Rose contradicts his post and says that 12,000-14,000 are submitted), there is too much stuff to filter through by yourself. To find meaningful stories you will either have to spend a lot of time digging through the upcoming section manually — or do what most everyone else does: Simply go to Digg’s front page.
If most people are only going to Digg’s front page and not seeking out new stories, that’s a problem. It means that still only a relatively small percentage are responsible for determining what eventually hits Digg’s main page. As a result, Digg front page has seemed somewhat stale and also redundant over the past several months.
By passively using other Digg users as filters for interesting news, we should see a lot more interesting content being dugg. If it works as planned, that system will perpetuate itself and Digg will be better for it. In theory, the more items you digg, the better the system will get at recommending new items to you.
This rollout comes at the right time for Digg as well. Competitors such as Mixx and Propeller (formerly Netscape) continue to grow (though they are both still much smaller than Digg). Meanwhile, arguably Digg’s biggest competitor, Reddit, recently open-sourced its code. This will not only allow developers to use all of their code as they see fit, it will also allow them to help improve Reddit.
Could we be about to see a resurgence of fresh, interesting content on Digg with this Recommendation Engine? This change is by far the most promising news in months. I’d digg it.
Watch the video below for more.
Digg Recommendation Engine from Kevin Rose on Vimeo.