XARK 3.0

  • Xark began as a group blog in June 2005 but continues today as founder Dan Conover's primary blog-home. Posts by longtime Xark authors Janet Edens and John Sloop may also appear alongside Dan's here from time to time, depending on whatever.

Xark media

  • ALIENS! SEX! MORE ALIENS! AND DUBYA, TOO! Handcrafted, xarky science fiction, lovingly typeset for your home printer!



Blog powered by Typepad
Member since 06/2005

Statcounter has my back

« 44 Minutes of Ed Cone | Main | Gotta be the Exams . . . . »

Thursday, April 24, 2008


Feed You can follow this conversation by subscribing to the comment feed for this post.



Considering my job at MedU is to manage clinical information flows, your post speaks volumes to me.

I want to take a minute to point out a problem I see with the upcoming "Semantic Web."

Your own foundational statement, "Because there are no unrelated topics," backs up the idea that humans relate things in ways that are totally alien to each other. This presents the problem:

How, when everyone sees things differently, are you going to be able to successfully create a set of unified and agreed upon semantics that encapsulates the unrelated/disorganized/unmaterial information that exists today?


Great question about the Semantic Web, and I think I've got a piece of the answer. It's an old concept called "Topic Maps," which really has very little to do with mapping.

It's basically the idea that your emphasize the RELATIONSHIP between bits of information rather than the TOPICS and categories we might assign to information.

Hence, a topic map for my family could include "documents" that represent me and Janet and David and Luke and Lee and Callie. But the topic map itself would be the "cloud" of relationship lines (spouse, parent, child, sibling, step-sibling, etc.) that represents the context of the family (i.e., relationships and the meanings that accompany them)

And that's just one topic map. The document "Dan" would have a different set of relational connectors in the topic map Xark, or the topic map P&C, or the topic map Carolinas bloggers, or the topic map fantasy football players.

If that just sounds like metadata, think of it this way: Metadata is contained within a document. But a topic map, with its relational annotations, exists BEYOND any metadata you might use to tag a document itself. The cloud is separate from the documents.

So I think we're going to rapidly approach the limits of folksonomies, and at that point we're really going to be interested in ideas like ZigTag, which attempts to bridge the gap between taxonomy and folksonomy. Is there a smart function that could create "standardized" tags based on known/previous tags? Mebbe.

But I suspect the transformational answer is going to be found in something closer to topic maps than in some document-based tagging approach. Human intelligence processes and retains information by putting it in stories and narratives. Topic mapping does something similar -- so that the same bit of data can have as many "meanings" as there are ways of using and understanding it.

Wanna go meta-meta? What happens when we get to the level where we're writing topic maps that describe the relationships between vast networks of topic maps?

Ralph Kramden
Everything is connected. Nothing is separate. I suspect it was always this way.

Absolutely. Any 30-second conversation with my Rabbi turns into 5 minutes easily and goes in directions completely unanticipated. His motto is "but I digress". :-) He has encyclopaedic knowledge and the wonderfully creative mind it takes to make these connections fluently.

The question is - where does that leave the (comparatively ungifted) rest of us? I need crutches, not only searches for information, but how that information interconnects, even in directions I may not have originally anticipated.

There is danger of some hubris here, though. In early '72 I took a class in Industrial Dynamics from Dennis and Donella Meadows, who had been part of a team that set out to model the world. At the basis of that model was the interconnections between all sorts of economic and social conditions. The methods were fascinating (to a math major), but the model was woefully incomplete, and the naive belief in the results were embarrassing, to say the least.

Can such a model be complete? I don't know, but I think Dennis still thinks so. But it comes back to the same point - everything is inter-related. And to go meta: even the existence of the model has to be factored in. (That much has been proven - one of the main focuses of models like these is the stock market, and once the models become prevalent, they affect the markets in big ways.)

Organizing information is a very old problem. It goes back at least as far as the Talmud, which starts with some sciptural text in the middle of a page, surrounded by commentary, which is in turn surrounded by commentary. The old Hypercard program on Macs in the late 80s was the first computer expression of "Hyperlinks". Lotus Notes had clickable links and tried to be a platform for organizing information, but it was blown away by the web.

Now, having a blog, even a blogging community, may not be the best way of having an "everything is connected" community. But it is a fine way of ordering discussion. Tags have to be made more visible and more standardized somehow.


I love that video.

I started my website as one of the "early" websites included in that video. Unfortunately, I have not done nearly so well in escaping the shelf. Paleothea (women in greek myths), still lists things alphabetically (shocking!!) within categories. And, again, it can be difficult to navigate. Sure, I have a search engine and an (embarrassingly incomplete) index, but I don't have the kind of jumping that a real understanding of webtechnology might give me. And, unless you sign up to have a real expert make that for you - which is more or less what happens when you get someone else to run a blog for you (like Wordpress, Livejournal, or Typepad) - that remains a struggle for most people. Especially because, sure, information is more available, but harder to connect to. Wikipedia is my favorite, because of it's depth and breadth, but what I love best about it is http://en.wikipedia.org/wiki/Main_Page - because it takes you places.

I want to take people places. I love the tours on places like Elfwood (http://www.elfwood.com/tours2/). But as was so aptly pointed out on Seth's Blog (http://sethgodin.typepad.com/seths_blog/2008/04/silly-traffic.html), it ain't easy to direct focused intent where it might get the best junk!

And relying on Google only rewards a certain kind of looking.

So. That doesn't have a clean ending. But nice post!

The comments to this entry are closed.