Is consuming social media content any different?

[former_member188228]

With all the empowerment and excitement social media tools bring for the individuals and enterprises, they also pose some uniquely interesting challenges when it comes to knowledge management. In this blog, I would like to call out a fundamental difference in which content driven applications can handle the social media content (content produced by the social media tools such as Blogs, Wiki, Forum, Twitter, etc) and the published content (the structured and unstructured content that is published for consumption of broader audiences via websites, shared file folders, document repositories, etc). The main difference is that – Unlike published content, consuming social media content almost always involves using the corresponding social media tools for interacting with the content as dynamic entities. This difference has certain ramifications on the design of the social media applications.

Social Media tools are used by the community members to connect with each other and collaboratively create content that is useful not just to the collaborating members but also to others who can access the content at some later point of time. For example, consumer sentiments regarding newly launched products can be tracked by monitoring the twitter space. Emerging trends and market opportunities can be spotted by taking a pulse of the blogosphere. Forum threads can be analyzed for extracting real input for sharpening the focus of solution management and improving the quality of solution support. In a nutshell, just like the published content, social media content has clearly become a great source of content that the search engines can grok and many other content driven applications can build upon. In the case of published content, the knowledge to be harnessed is embedded in the content itself which the consuming applications can parse and extract, or in the simplest scenario the knowledge is rendered to the end user via a reader/player tool corresponding to the content-type of the published content e.g. acrobat reader, word, video player, etc. However in the case of the social media content, the object of interest is a combination of the content itself and the collaboration instance associated with that content e.g. consuming a forum thread can be fully realized by enabling not just access to the forum content but also by supporting participation in the thread activity such as for replying to the discussion points in the thread, rating the responses, etc. Similarly, consuming a blog means accessing the blog content as well as potentially commenting on the blog, and likewise consuming a wiki page may involve applying some edits to the wiki page itself.

From a practical standpoint, the above difference manifests in various constrains as well as new capabilities such as the ones listed below.

Consider the simple case of a search application. When the social media content is included in the search results, it wouldn’t make much sense to include just the cached copies of the snippets of the social media content – as hinted above the consumption of social media content would typically involve interaction with and evolution of the content.

Rich content-driven applications cannot apply a single brute force approach to suck up and process the various kinds of social media content. Instead, such applications need to become cognizant of the social media tool specific mechanisms, standard or customary, for contextualizing and processing the content. For example, to meaningful process a Forum thread would mean understanding the conversational structure of the content, identifying whether the thread was successful in answering the original question, etc. Similarly, in the case of a twitter message, all the messages sharing the same #hashtag may need to be processed to grasp the full context of any given tweet (assuming there is a hashtag for that tweet). In essence, applications that leverage the social media content will have to build upon the unique collaboration models of the individual social media tools, which would typically require a close integration with the tools and support for the corresponding interaction patterns, etc. Depending upon whether the tool is deployed in the enterprise space or it is out there on the net, etc, standardized APIs and protocols (RSS feeds, etc) may get utilized.

Unlike in the case of published content, applications processing social media content also need to pay close attention to the paradigm of community-member-is-the-center-of-the-universe. When a community member makes a certain contribution, it also results in a certain enrichment of the member’s profile and reputation in the community. The social media applications may involve analyzing the content as well as identifying the right experts in that area. Although this scenario of identifying the subject matter experts is technically valid in the case of published content, it is much more natural for social media scenarios where the members, their contributions, the resulting content, the collaboration, etc, are all intricately woven together in one single ecosystem (of which there may be many instances). Also, the social media applications (such as searching and aggregating the related social media content) also need to employ user-centric design principles in delivering its own user experience. For example, instead of providing a non-intuitive navigation scheme for displaying the search results, a well designed search application can deliver the search results by taking into consideration the choices and preferences of the user issuing the search query.

Another unique aspect of the social media content is regarding the assumptions one can make about the availability and accuracy of the semantic metadata describing the content. For social media content, it is much easier and natural (compared with the case for published content) to include certain steps in the social media interactions whereby the community members can readily provide the metadata for semantic description of the content. Whether it is free-form user tagging or whether some smart mechanism is deployed for providing the community members with a choice of governed tags, it would be reasonable for the social media applications to assume that the social media content is associated with some semantic tags. Making similar assumptions for published content does not seem realistic.

The diversity of social media tools is most likely going to be a reality to deal with for a long time. Efforts such as Google Wave may provide a consolidated platform for some of the social media capabilities, but in the bigger picture, Wave may be adding yet another tool to work with, at least until the dust settles and clear winners stand out. It will therefore be important to understand the unique characteristics of the social media content so that patterns, techniques and standards can slowly emerge to make the consumption of the variety of social media content from other mainstream applications much more efficient.