There are hundreds of millions of people using social media sites like Facebook, Twitter, Digg, and many more. For the most part, these sites have not changed significantly. Some sites, like Facebook and FriendFeed, have modified their display to react quicker to the “real time web”. Others have changed so little, like Twitter, that mini-industries or ecosystems have been created to extend their functionality.

Social media has gone mainstream, and nobody can deny it. Oprah is on Twitter and Facebook is used by over 300 million people. The mass consumer has found their old friends and has a voice in the public unlike what was possible before. We have seen political success in the Obama campaign and even political uprising in the Iran election, all on social media.

Now, the question is, what is next? Or even, now that we are on social media, what do we do now? From the mass consumer perspective, nothing really needs to change yet. Smart people are trying to develop the next big thing in social media, and early adopters will try to anoint one as king long before it gets to the mass consumer. So, what about the early adopters and businesses? That is where things need to change or at least improve.

Social Media Monitoring and Searching

Now that so much is being said on the various social media sites, it is logical that monitoring and searching these sites from a unified interface is critical. If you read this blog regularly, you now that I have a significant interest in the social media monitoring space and have created a monitoring tool in YackTrack. There are various free and paid tools to help you monitor social media. There are also several “real-time” search engines, with more being announced every day. However social media monitoring and searching is still very immature. Sure, there are trend graphs and email alerts, but should there be more than that? For the monitoring tools, what features are needed to really make monitoring a common occurrence? For real time search, what problem are the tools really trying to solve? Google has a very good handle on regular search, and it is only a matter of time before they add real time streams to their indexes. What makes real time search a necessity? If we answer that question, the business models will quickly follow.

Content Discovery, Filtering and Curation

The amount of content shared on the various social sites is huge. However, many users cannot really find the content that they want, or they do not know how to deal with the information overload. There is also the problem of where good content may be found that is not from a major blog or news source. How does a blog really get discovered? Part of the model for social news sites is the curation of content in the form of voting. As effective as voting is, it is hampered by the fact that major blogs and news sites have a huge advantage by being shared often, and the “blind voting” issues that go with it. Are those posts the best articles available? Not all of the time. To avoid this problem, filtering and more human curation were added to different sites. Mixx has their communities and SocialMedian used newsmakers, networks and topic filters. You can also use TechMeme as an example of human-aided curation.

However, with all of these options, people continue to complain that they cannot find content relevant to their interests. Digg is notorious for their “power Diggers”, where if you do not have a power Digger’s support, you will likely not reach the front page. People have even complained that TechMeme plays favorites with some sites. Content filters or topics have always been only slightly helpful in many cases as well. So, how do we get around these problems? Will semantic technology help? Will the linked data initiatives provide the level of metadata we require to intelligently and automatically process all of this content? Right now the answer is no, but there is a lot of research in semantics and linked data, so it could change quickly.

Stability, Federation and Distribution

A social media post would not be complete without mentioning and complaining about Twitter and Facebook. There are a few concerns currently with sites like Twitter and Facebook. First, they have a huge amount of traffic flowing through them, and they have become “defacto standards”. The problem with this is that they have appeared unstable at times. Another problem is the typical dislike for monolithic platforms. There have been many calls for a federated or distributed Twitter, and I have written one myself.  The idea behind this is to have Twitter become infrastructure behind many applications. This allows Twitter to work much like email does, where Twitter itself may not need to own the servers. By having a distributed (or federated) network of servers, the perceived stability of the service improves. By having a more stable solution, people will not be as concerned about relying on Twitter for business. The same holds true for Facebook. There may not have been as many outages when compared to Twitter, but there is still the concern of a monolithic system. For microblogging, there are other solutions available like the Laconi.ca installations or even the hosted version in Status.net. Social networking has not been approached in the same manner. Initiatives like OpenID exist for open identity solutions, and OpenSocial exists for “social widgets”. However, these are just parts of the social networking landscape.

Building Blocks

If you look closely at the changes that have slowly taken shape over the past year, you will realize that we were in a foundation phase. Social media sites have been laying the basic groundwork for the future. We have seen the ecosystem around Twitter grow. We have seen applications built upon Facebook. OpenSocial and FacebookConnect have allowed us to bring our social networks outside of the initial application. The foundation has been given to us, and we need to start building the next iteration of the web. Is the next iteration the “real-time” or “2010” web? No. Is the next iteration the much anticipated Semantic Web? No. Is it some weird mashup of features that we likely would not have thought we needed? Probably. Of course, the question remains, what is missing?