Several years ago, while teaching a graduate course in distributed computing and peer-to-peer architectures, I was inspired to develop a collaborative, deliberative discourse system that would be free of centralized servers and scale globally. To that end I developed a concept I call ConsensUs. You can find the base (still rough draft) document at: http://faculty.washington.edu/gmobus/ConsensUs.html
Note in particular the design notions of using several AI approaches to keep the content manageable for users.
I've been stalled on this project due to lack of talented graduate students (we're a primarily undergraduate branch of the Seattle UW campus). I look forward to sharing ideas about how a community of interest can further the implementation for some form of global-scale discourse.
In response to your question if I think ConsensUs is worth further pursuit, my answer is a strong affirmative, with the following caveat: there's been a tremendous amount of water flowing under that bridge since you last wrote about it and I'd like to see how you factor all the recent developments into your thinking. At the very least, there may be other projects that could attract your enthusiasm and energy.
I'm not sure just how much collective energy there really is in this game of crafting portals to save the planet; I'm noticing that most everybody with whom I've spoken is already following some path or other and really they don't have time or energy to contribute to other projects. If we were to create something like a "sensemaking forge" and, perhaps, attract some funds that would support, say, a competitive analysis and exercise of many of these projects, perhaps we would begin the process of sensemaking the sensemaking industry. Just a thought. Your mileage might vary.
Does the topic tree impose too much structure? Perhaps there is no support for an answer either way, but my own intuition, as is being expressed in my thesis project is that the topic tree is akin to the "secret sauce" that will help hypermedia discourse be more effective. I call it a topic map.
In my day job, we built a desktop platform called CALO that uses both LSA and LDA (topic modeling) to tease out of text resources important bits of focused information. We managed to put most of the cool stuff into an open source platform: OpenIRIS. I say that to respond to interest in using such methods in text mining. They remain important, but they do not solve everything. An even more interesting approach uses combinations of text tagging and concept mapping to harvest text. You can play with a search engine that uses the TextRunner algorithm here, and also find links to papers about it. I believe that the TextRunner algorithm can be improved in accuracy by shifting from a concept map to a topic map.
I used the term "sensemaking forge" in the same way we would talk about sourceforge.net; a portal that uses sensemaking tools of hypermedia discourse in support of the creation of sensemaking tools. When I built my first topic map/wiki platform, I noticed that I could write the source code in the wiki itself and use the topic map component to wire various methods, class variables, etc, to other discussions such as use cases, etc. Now, we know we can tag other websites offering algorithms, code, etc, to bits of code, use cases, requirements, whatever. The use of our tools in software development could mark a paradigm shift, perhaps the equivalent of the move from ISO methods to Extreme Programming. Perhaps.
My own topic map is online at http://www.topicspaces.org:8080/wiki/
I don't want that URL loose in google; the platform is running on a very slow DSL in my bedroom. Be kind to it. What you would see if you visit is, essentially, nothing. The front page greets you with an ajaxy drill-down navigator that sometimes works (have some css bugs still to remove). Ignore that. Visit Tagomizer, visit Blogs, visit other things including the Help and Feedback pages.
The ConsensUs project, in my mind, remains valuable.
I was in high hopes when I first discovered the GSm, because of the similarities of my ambitions with the Do Good Gauge. Before the GSm, I customized over one hundred letters to various authors, universities / professors, politicians, and journalist over a 8 month period. The approach got me about 10% response and about 2% meaningful but short lived dialog.
My initial posts and responses to the GSm Ning were promising. An extended email exchange with Jack Park was a great learning experience for myself and I would hope provided some benefits for Jack.
Since then, I have found the GSm ning to lack direction, focus, or provide feedback.
Though I claim the requirement for developing a process for creating an intelligent argument is not necessary for the Do Good Gauge, I have thought much about this requirement. Specifically, I see Benjamin Franklin's Junto as an intelligent argument development process which has rarely been duplicated with such success and longevity. As a member of the House of Junto Google group, I initiated some thoughts about Benjamin Franklin's success with his Junto:
As a group, I feel the GSm is too focused on tools which granularize argument into such small grains that the problem is lost on the beach.
As a pragmatic software engineer, I believe it is better to develop software based on a working manual process. As a suggestion, I would recommend forming a Junto based on Benjamin Franklin's principles. Instead of focusing on the GSm tools, I would suggest this group collaborate to develop several intelligent arguments of publication quality. I would focus on narrow topics. An example could be, a patient managed medical record repository. I started a thread on the House of Junto to discuss this topic.
The lessons learned by manually creating intelligent arguments could be beneficial to generating tools. Since the Do Good Gauge does not require the toolset for generating an intelligent argument, these manually generated arguments could be used as an inspiration for developing the Do Good Gauge framework.
George, I was hoping that your "Philosophical Aspect of Sense Making" topic would interest you in continuing this dialog. I would be glad to discuss with you on or off line.
Just joined. I like learning stuff too. I also have a long-standing passion for dialogue, our earth, our wicked problems v a more peaceful planet, and a congruent walk trough this life. Plus I have a couple of old visions for tools that could only come to life in cyberspace (is that term still used?), and this generates curiosity in computer apps. I'm a systems thinker by nature and a resource developer by profession and practice, as well as a writer and all those other things I gave in my bio. Hope that this net will be as fascinating and productive as my hunch tells me it will be. Lynne