One of my active research areas right now involves event stream processing and its implications to communication, collaboration and social networking. Specifically, I'm looking at how attention data and "post" activities act as informal, loosely-coupled signaling methods that, when streamed in a public manner, can be combined with sensor/filter/relay mechanisms to intelligently pull messages and information to other people or situate the information to the right place (e.g., a "my space" created as a honey pot of sorts to house interesting/relevant items). A goal of such an environment would be to create opportunities for subsequent connection and interaction (at a person-to-person, group, team, or community level). Examples of attention data and post events would include blog entries, tags, social bookmarks, and changes to presence indicators - even Twitter messages are a continuous stream of post events. The collection and streaming aspects could be accomplished through a combination of pinging, aggregation and XML syndication services. The sensor/filter/relay mechanisms could be instantiated through various clients and agent/watchers. This type of environment would likely not be a single system (overly complex and monolithic) but instead would be some set of inter-connected / net-centric systems of interoperating nodes, without hierarchy, and include some level of trust zones (for security/identity reasons).
Why look into this? Years ago I became fascinated by what DISA and DoD were doing around net-centric collaboration. The DoD felt that it was imperative to move from Task, Process, Exploit, Disseminate (TPED) to Task, Post, Process, Use (TPPU). The design emphasis is on posting, processing and using data and information in parallel. A key aspect was enabling processing to happen in parallel to information access with people having access to information refinement along the way. The thinking (I believe) is that getting information out to those that have a "right to know" through an intelligent pull model is more effective than pushing information through a process first (introducing latency) that makes assumptions about who "needs to know" (often driven by a desire for risk avoidance).
A TPPU model is about managing risk while providing near-instantaneous communication and information access. It avoids scenarios where people are not alerted and failed to become knowledgeable about situations that impact them. It is almost impossible to know who might have a time-sensitive or other relevant interest in the raw data and its refinement as the process cycle continues. It was felt that a TPPU approach would reduce communication latency and improve information transparency. The result of such an environment would be improved shared situational awareness and self-synchronization through net-centric collaboration. This would have obvious operational and battlefield benefits and have a transformational impact on future warfare strategies.
Adopting and applying these concepts more broadly is intriguing. Almost all organizations could benefit from a TPPU-like model in terms of improved enterprise agility, resiliency, and performance - and would positively influence innovation and knowledge management efforts). This, in a round-about way, gets back to "the need to know" vs. "the right to know". It does not necessarily mean an end to push-centric systems, they continue along in parallel to intelligent pull models. Clarifying my thinking around the balance between the two models was triggered by a comment from Urban Mermaid below. It's a fair point and highlighted some incomplete "thinking out loud" on my part in an earlier blog entry.
In any case, this post provides a glimpse into some current research activities. Feel free to comment ... this is not fully-baked in my mind, I'm bouncing ideas off of various clients and vendors and found this post by Jeff Jonas especially helpful.
Comment from Urban Mermaid
interesting that you mention people wading through the stream of information and extracting what they want. however, what i've found in my job of providing research & analysis is that i still have to actively push information to people.
it's like the advertising mandate to break through the clutter.
sure, there's definitely people willing to wade through it and that's a cultural value that's beginning to embed itself, but i also feel there's still a need to have some information that gets fed to people.
I couldn't get the trackback to work, but I did have a comment about this post:
http://developer.xythos.com/files/tgrant/blog/blog.html
Posted by: Tom Grant | April 02, 2007 at 02:37 PM