Enabling Sensemaking via Social Networking Sockets

Replies to This Discussion

Permalink Reply by Mark Szpakowski on May 21, 2008 at 19:45

A plus here is that Ning supports OpenSocial and, soon, Friend Connect.

Permalink Reply by Jack Park on May 21, 2008 at 23:40

Were I to read that post in an IBIS map, I'd be inclined to raise a Con node with the justification (my own personal biases showing here) that Google makes no bones about the idea of running OpenSocial for purposes of improving their ability to target advertisements at me. I'll admit that there is merit in reducing the amount of noise in the ads I must already face with them, but there is something about them "watching me" closely that stands my hair up. I gather that Friend Connect is also a Google product. Haven't read anything about it yet.

I guess that what I am arguing for in most posts I make around here is for an environment that is the least imposing on the sensemakers and students who will join. I'm pretty certain that some people, say, 4-star types in the military, high visibility people like politicians, and others are not going to want to lay tracks in social sensemaking networks that so blatantly watch their every move. I already have strong evidence that Google's desktop tools are not accepted as components of the sensemaking tools my employer has delivered.

In the long run, I might argue the opposite side of your intuitions, that we do need a myriad of small sensemaking components 'out there' that we can tie together not with an IP registry that tracks surfing behavior but instead with at topic map that organizes the statements, questions, and arguments, the stories, in a subject-centric way, that we do not need the likes of OpenSocial or Friend Connect. We can always toss in FOAF and many of the other emerging social semantic web microformats and tools. In fact, we can always make our work products available as RDF snippets to any website that wants them. FWIW, I see three interesting (if only to me) trends:

1- the migration from document-centric computing to subject-centric computing
2- the migration to social discourse, which, in my view, is just about to realize the potential mistake of treating humans like documents and is about to realize that it is the subjects those humans hold in common that represents the real commodity
3- the slowly dawning realization--perhaps driven by the "emergence" of wicked problems (climate, energy, etc) --that thinking above the plain of profits is more important than thinking first about the profits.

Permalink Reply by Andy Streich on May 22, 2008 at 0:14

"a dynamic representation of a Sensemaking tool's work output" -- that's an important concept to me. What are the characteristics of a Sensemaking tool's work output?

Seems to me we're faced with the fundamental problem of deciding on a knowledge representation scheme in the absence of any well-defined criteria. Or am I off in left field?

Permalink Reply by Jack Park on May 22, 2008 at 0:31

My view is that it makes sense to allow, maybe encourage each aspect of sensemaking to work out its own representation schemes. Compendium, for instance, has an XML serialization that defines nodes and arcs, with attributes that represent dates, display location and other items, together with containers for the title string and details string. I'm not certain what goes on inside Cohere; what's more, I don't think I need to know. What I do need to know is how it serializes its work product to support web services queries: how it's web services protocols are defined. Compendium does not (at this time) offer a web services interface (I bet that's somewhere on the Todo list). So, arriving at mutually acceptable web services protocols makes sense to me.

Off in left field? I don't think so. If you give me access to your serialization scheme, I can write a translator that turns your work product into an artifact usable in my topic map. That's what I care about. And, I'll offer you a serialization and protocol that you ought to be able to use if you need one. The topic map is there to federate the work products of a variety of sensemaking tools. I'd like to think that most all of the various tools will arrive at as close to "standard" web services protocols as possible. I've had no problem cloning the protocols of http://del.icio.us/ for Tagomizer. I can already read and write those of Cohere.

Permalink Reply by Mark Szpakowski on May 22, 2008 at 16:39

Andy and Jack, I actually meant something much more simple-minded than publishing either a schema or a subject-mappable file. I was thinking more of representation as in REST, where the publishing site returns some dynamically generated (so it's up to date) image (at worst) or interactive rectangle (through a widget or other API-driven ways) to be rendered on the calling site. Just as here we get a representation of the Debategraph map, with some interactivity for opening a Debategraph-hosted version of the map into which I can log in.

Friend/Facebook/*- Connect, or equivalent data/social-graph portability API, would enable transparent moving of this Ning group to Debategraph and back.

The motiviation is to make it as dead-simple as possible for an existing group to access these new Sensemaking capabilities without requiring them to rebuild their entire social network of people and permissioned resources on another site and then have to keep two sites in synch. The group can stay where it is, but be able to extend its reach both to create new (Sensemaking) content and to embed it in their site.

Permalink Reply by Jack Park on May 22, 2008 at 18:09

Let me toss out an interpretation: transclude tools and content through widgets such that anybody can bring sensemaking to their home page by way of a simple REST interaction. If that is even close, then I would point out that Tagomizer/del.icio.us do a part of that by giving you a REST-like bookmarklet button so that no matter where you are on the web, you're wired to Tagomizer/del.icio.us and TopicSpaces and similar types of sites, e.g. DebateGraph already let you use a REST-like URL to fetch resources to paint in as simple a widget as an iframe. We could imagine extending that to some imagined library of javascript widgets coupled with REST URLs to do even more. If the interpretation is what is suggested as a dead-simple as possible methodology, then are we not aimed already in that direction?

Permalink Reply by Jack Park on May 24, 2008 at 5:40

Mark, I'd like to play with this thread for a moment. My suspicion is that your vision sees things at the quantum level while I'm still out here in the clouds trying to see molecules. If I'm reading your statements anywhere near what you are thinking, then I'm seeing something along the lines of individual web services servers that do small, simple things like, say DebateGraph if we restrict our needs to something like just letting us create and manipulate IBIS maps, or maybe Cohere, which allows us to send it things that we then go in and wire up with coherence relations, all of which we can bring back to our site when someone wants to roam our site. "Our site" would be, say, a climate site, or maybe some subset such as hurricanes--your practitioners. That way, anybody could toss up a CMS that does blogging or something else, and just authenticate with various services and compose their own sensemaking site. Am I close?