Search engineer here.
In the video, the commenter admits the proposed algorithm works by retrieving 10 pages worth of documents and then analyzing all of them for possible social data. This is a terrible solution, since it significantly decreases query time performance. You might say, "Well, then let's move this code to index time and slow down THAT process instead! It's less important, right?" I'll have to disagree here for a number of reasons. The most obvious one being that social media changes so frequently that relying on this method will cripple the importance of results.
Let's remind ourselves that Google has tried multiple times to add a social touch to their results. This started with the realtime feed from Twitter that ended because the Twitter/Google deal went sour.
Let's also not forget that Google has tried a number of times to integrate better with social platforms. They've basically waged a war with Facebook over allowing customers to migrate data more freely between services.
One more point: This video spends a lot of time talking about finding people on social services through Google's search. The purpose of SPYW isn't just to find people. The search bar in Facebook, G+, Twitter, and every other service can already do that for you.
I'm not saying Google's perfect here. I'm just saying this proposed solution is terrible and the slant these companies are trying to add to this discussion isn't accurate.
The best of ZDNet, delivered
ZDNet Newsletters
Get the best of ZDNet delivered straight to your inbox
