Last year I started looking at ways to handle the amount of Social Media presence in a reasonable way. Then, I came up with a Social Media Interconnection Map and a set of design rules, principles and guidelines that we could take into account shall we wanted to build our own.
Unfortunately, everything moves insanely fast in the technology space and so many things have changed since that first proposal. In fact, I have had to adapt my original design several times over the past months due to a number of reasons:
- new players have emerged or exploded: Google+, Pinterest, Tumblr, Posterous, Instagram, SlideShare, LinksAlpha, ifttt, dlvr.it, So.cl, Outlook.com, …
- new types of services have also appeared: Google Hangouts, Google Events, SocialBro, BufferApp, …
- some services have disappeared: Google Buzz, PicPlz, LightBox, Jaiku, ping.fm, …
- … and others have become less relevant: blip.tv, status.net, …
- some APIs have changed and/or so do its Terms and Conditions: Twitter not playing with LinkedIn anymore and breaking the 140 character limit with the notion of “expanded tweets”; Google’s new Terms & Conditions; the “new” write access to Google+ Pages; the Facebook Open Graph adoption; Spotify integration with Facebook; …
- relevant Mergers & Acquisitions have taken place: SlideShare is now owned by LinkedIn, Posterous and TweetDeck have been acquired by Twitter, Yammer and Skype by Microsoft; Instagram now belongs to Facebook, Trunk.ly to delicious.com and Radian6 to SalesForce.com; SocialCast is now part of VMWare’s Portfolio and so did Citrix with Podio, Atos with BlueKiwi, etc…
- the availability of resilient, reliable and easy to use filtering and transformation tools. Though Yahoo Pipes has always been there. However, ifttt and dlvr.it, for example, are much more effective for basic use cases.
- being able to deliver relevant and curated information inside your organization in an automated way is, without a doubt, more than interesting. For example, pushing customized information feeds into your corporate SharePoint, or any other ECM or ESN system make it easier to build and nurture internal communities.
- the original hubs and/or distribution channels -ping.fm and FriendFeed– have proven to be unreliable and it was necessary to find other alternatives for them. For example: ifttt, LinksAlpha, BufferApp, dlvr.it or HootSuite could be some of them.
- experience has shown that too many stages in the pipeline of the Interconnection Map increase the probability of error. So, new designs should try to minimize this risk.
- experience has also demonstrated that chaining URL shortening services over the different steps of the publishing pipeline can lead to failures or unreliable behavior. So, careful planning is necessary on this point: ow.ly is used in HootSuite; t.co is enforced in Twitter; wp.me might be used on your blog; bitly.com might be used in ifttt, dlvr.it, BufferApp, …
- engaging and working with groups and subgroups inside each social network has become increasingly important: Google+ circles, Twitter Lists, Facebook Lists, Facebook Groups, LinkedIn Groups, Yammer Groups, etc.
- last, but not least, engaging with audiences in different time zones or attention windows is, more than ever, a requirement. Therefore tools for content scheduling and looping are necessary: SeesMic, HootSuite, dlvr.it, BufferApp, SocialBro, etc.
Redefining the pipeline …
With this changes and new requirements in mind, we can now revisit the definition of the different stages and roles that compose our original content pipeline:
- Content Generators: services that allow us to publish our content and/or activity as an RSS/Atom feed or through their own open APIs.
- Hubs: services where we can either aggregate RSS/Atom feeds from foreign services or leverage adapters to consume content and/or activity from them through their open APIs.
- Filters: services that allow us to declare basic content selection criteria and enable provide content transformation capabilities.
- Distributors: services that allow us to republish content on different Target Services either as RSS/Atom feeds or through adapters that consume their open APIs.
- Target Services: interactive communities where we want to share or publish our content.
Therefore, the newly “updated” Content Pipeline would look like this …
… and the a new new Social Media Interconnection Map could look something like this:
So, what can we learn from this “creative exercise”?
First and foremost, the whole ecosystem is still boiling with change and evolution so new services appear and the existing ones grow in sophistication, reliability and completeness. Some others, on the other side, just vanish or get acquired. The engagement and content consumption patterns, themselves, also evolve and introduce their own degree of entropy. In other words, the overall complexity is growing and changing and will continue to do so at significant rates.
If that is the context, mastering the “art of the mashup” will become increasingly important. If you compare how was the last configuration of the Interconnection Map and compare it with this one, it is fairly easy to make oneself an idea. In fact, this new “skill” could be key to make a difference in the Cloud and the Social Media ecosystem for certain business activities.
There is no “right answer”. Actually, there are as many Interconnection Maps as goals we can imagine. Hopefully, this new service mashup can help you set up your own. Of course, you can mimic this one if you want to. However, one thing looks pretty sure to me: we will see another version next year. Any ideas about future layouts? And, more importantly, what do you think about this particular subject?
