Joho the Blog
An Entry from the Archives

I wonder if Clay Shirky was in the circle and if you have substantial agreement or dis- regarding ontology?

I wonder if tags might not be less interesting than the label "folksonomy" makes them? Matt Mullenweg said, "I’ve had a couple of people ask how best to tag me in a photo, I’ve seen some use “mullenweg” or “matt” or “mattmullenweg” but the best bet (and the one I watch) is “photomatt“."

The researcher who MUST assemble snapshots of Matt will - if her IQ is anywhere above room temperature - infer these tags without Matt's direction. Over his lifetime, a long one I hope, there may be a convergence to "photomatt" if he becomes THE photomatt. There's another photomatt (Matthew Gilbert) who owns the .com URI. Mullenweg is the .net guy. Or, who knows, maybe the folks of 2020 will tag him as WordPresser... or something. The point is, this is all soft and loose and the discipline of semantic structural analysis and kraftwerk leading therefrom isn't something that the pop-techny community is putting a lot of energy into. Rather, I think we'll see the important infrastructure emerge from places like Cern and the Large Hadron Collider people or the Internet 2 effort or some serious (gag) corporate R&D. Our little virtual village grows incrementally with cool ideas like del.icio.us, but our virtual world is built on a well planned framework brought to us by big capital... the foundation pieces with huge price tags include things like Sprint's TAT 14 Cable System, Dense Wavelength Division Multiplexing, all that hoot.

I look around this office and see maybe 300 books. A couple of shelves are full of binders that each contain multiple datasets, but by and large the books are single titles and depending on where my work takes me I may need to draw on three or four of those titles to assemble info on a variety of subjects. My only clue as to what's on these shelves is that I thought when I shelved the book that it was one I'd like to have near me. My desks and bookshelves prove that everything is miscellaneous. I'm eager to read your book about that David.

Frank, I agree with Clay to a distressing degree. His talk on the topic at eTech was brilliant, IMO, and I even discovered that(through the power of selective forgetting) I lifted the Dewey example I've been using from a talk I heard him give a couple of years ago.

Tagging would be happening without the happy "folksonomy" term, imo, because some apps give us good individual reasons to tag, which then turns out to have quite spectacular social effects. And I disagree that the tagging app providers aren't thinking about the hard issues that arise when you have millions of people using short strings to tag millions of objects. They are, which holds out hope that the solutions will be nicely tied to actual apps as opposed to being imposed context-free in a Dewey-like manner. Plus, we'll figure out a bunch of stuff bottom-up, just as in your photomatt example.

I think.

I was curious if you've encountered the work of Gilles Deleuze. He's a very French philosopher, died in the 80s, but he's pretty huge over there. I focus on him in studies.
His most famous book, co-authored with Felix Guattari, is called A Thousand Plateaus. The first chapter is called "Introduction: The Rhizome." They contrast the mode of organizing beings and ideas in a "tree" form (based around, say, Platonism or abstract ideals tightly ordered by rules and codes) with organizating according to a "rhizome" form (forming horizontal bonds across very different beings that allows them to create new beings/concepts).
Your argument about organizing knowledge as trees sounds very similar to the implications they're getting at, and it made me wonder if there are any connections between your work and theirs. If not, you might like reading them. Like I said, very French, but ultimately very rewarding.

Manuel Delanda is a very good commentator on their work with several essays online. This one is very good:
"Meshworks, Hierarchies, and Interfaces"
http://t0.or.at/delanda/meshwork.htm

"Ask why tagging is happening now?" ... Well this might be the first time that we have tools where tagging is not just lonely and tedious. The first time we can point to something and say *in a social manner* this on one of those. Our tools are finally working the way we learn language.

I know that Pattie Maes of MIT labs is doing some research with 'coRelate Networks' /Social networks.
======
"CoRelate Networks has mined online social network communities to create a rich influence network of interests and subcultures. Some of the interests represented in the network are, inter alia, television and films, foods, geographies, music, sports, hobbies, activities, objects, and people. The strength of connections between interests are learned from the digestion of on the order of one hundred thousand user profiles from online social networks. For example, a person who likes X may also mention Y as an interest on her homepage. Identity and tastes emerge as patterns of intersection on a coRelate network"
=======

Dave's, post on taxonomy and tags has really sparked some my interest's back again to KBE, smart agents and Knowledge Semantics. Tagging may just be the begining of true knowledge mangement in digital form for the global digital village. I am thinking that tagging is one portion of the puzzle. Today, we have tagging. Tomorrow, we'll have interest engines. The day after, we'll have smart agents, that corelate interests and tags and derive at meaning to that indivual user.. who knows .. what the social tapestry will look like in the future.. !!

"Today, we have tagging. Tomorrow, we'll have interest engines. The day after, we'll have smart agents, that corelate interests and tags and derive at meaning to that indivual user.. who knows .. what the social tapestry will look like in the future.. !!"

Well said, Peter. I'm looking forward to the day when we don't need to go through the rigamarole of making a group to discuss a topic, nor find some domain in a haystack where it is being dicussed by the small group who happened to find that discussion .... rather you we can just start posting to the topic and those interested will see it and respond.

David this is a very interesting thread. Organizing information into trees looks good and works reasonably well to start, but falls down quickly when the tree gets reasonably large. You can spend more time wandering around the tree than is good for you.

Better techniques are indeed needed to organinze, manage and locate information. I've recently had someone comment to me that adding keywords to content and then searching based solely on those keywords enables him to find any information he is after far quicker than other methods. I have yet to try this myself, but instinctively it sounds good and I'm looking forward giving it a go in Surfulater.