Social Rank v. Page Rank - the more things change, the more things stay the same.
I was reading this piece on Mashable, by the ever thought-provoking Steven Hodson. And true to form, my thoughts were provoked.
In that piece he talks about the importance of Social Media as the means to discovering top content. which seems to jibe with Louis Grays post about how the primary traffic drivers to his site are now the social media sites like reddit, friendfeed and twitter and the big blogs like techcrunch are much less important.
Something about the article was making my spidey sense tingle, though, and I’ve been thinking about it. Steve suggests that this is key:
When you have startups like Gnip looking to be the gateway between social media content producers and the consumers of all that information you can begin to see that they would be able to quantify who is getting all the link love on a realtime basis, plus they could create an incredible meta database of content around which any ranking system could be applied against. Where Google updates its PageRank values when they feel like it, something like this Social Media Index would be a realtime beast that changes with every Twitter posting, every blog posting, every FriendFeed posting or any of the many other social media providers or comments made via third party commenting platforms like Disqus.
To me, though, it seems like this is just changing PageRank to use, essentially, a subset of the information out there. It doesn’t fundamentally seem to be different, more or less an accounting of links by folks with varying degrees of importance. On the one hand, a difference is that Gnip would have it’s data pushed to it where Google has to go and get it. But relying on an opt in service like Gnip would provide a very self-selected audience of data providers- Google’s is much more, for lack of a better word, inclusive. It goes out and gets everyone.
Another difference would be the real time nature of this. But at the moment, that’s simply a matter of scale. Should this theoretical social media website get to significant scale, it would almost certainly not be able to provide real time information having to compute this on some periodic basis, just like almost every massive recommendation engine is computed, see netflix or amazon rankings. As an inverse example, look at Google News, here Google’s chopped down it’s working set to a much more manageable (if still quite large) pool of pages and it updates very, very quickly.
Another problem with isolating down to social media for the general case is that it’s very bursty and about what’s hot now. This is, in fact, one of it’s strengths - information finds you because everyone’s chatting about it. But it’s also not necessarily great for a general purpose search engine looking for a huge variety of evergreen topics and what not.
It’s interesting because he closes here:
Where once we had something like Google’s PageRank dictating what is popular, something like a Social Media Index would change the game by letting the people decide what is popular.
It was not long ago, that blogging was the people. PageRank is, to a huge extent, driven by blogging. Look at the effects of google bombing, the wide spread use of Pay-per-Post, etc… The number of blogs in the world and their propensity for outlinking to a large extent drives PR. Now there’s a subtle shift, with the advent of micro-blogging (like twitter), social recommendation engines (like reddit) and aggregated comments (like Disqus) amongst many others there’s an even lower barrier to entry to publishing on the web allowing orders of magnitude more people to get in on it than before, just like what blogging did when it hit the scene.
From my perspective there’s two options. One is, essentially, an extension of PageRank to take into account these new services. I suspect, in some ways, that will happen automatically as Google simply indexes the permalinks for each of these micro-publications. Perhaps a startup will get in on this and do it better, but if the database includes *only* those social media sites I don’t think I’d switch off Google for my general purpose searching in much the same way I didn’t decide to switch to technorati for search when it burst on the scene (nor google blog search for that matter), even before spam blogs became a huge problem (and believe me, as social media gains more importance in mainstream content discovery, be ready for the spammers and assorted black hat friends).
On the other hand, for those willing to put the time into developing a community of voices they trust, they’ll have created their own little web that can be focused and provide much more targetted content. (talked about this earlier this year).
The advantage of this is obvious, trusted content sources and very difficult to game, for a provided topic this would be difficult to beat, it’s already possible now and will only get better. The disadvantages lie in the very focused, echo-chambery notion of always selecting the voices you listen to as well as the limited type of content that will be surfaced. In addition, I suspect, very few will go to the trouble of developing that circle of experts. On the other hand, the ability to publically publish that circle, could be a very interesting development.
There’s no doubt in my mind that this is both a good happening and that it will have a serious and long term effect on the internet. By extension it will also have an effect on how people discover content. But again, it seems very evolutionary and a step that Google might be doing almost by default, as it crawls public twitter pages. On the other hand, it’s a torrent of information as more and more people get into the action. We’ll see what happens, but it looks a lot to me like this is just history repeating itself. What do you think? Giant step forward? Or we still going along at the same velocity?
