Using to discard stodgy taxonomies and enable the semantic web

Several months ago, somebody introduced me to  It was positioned to me as a "bookmark manager", a label that once meant frivolous dotcom business idea to me.  And then Jon Udell began singing its praises which is frequently an indicator of something important.  On a recent visit at the InfoWorld headquarters, Jon showed several of us once again why this type of product matters.  I still didn't get it.

Over the weekend, however, while the baby was sleeping in the early Sunday morning hours, I ventured into and began playing with some ideas, applying them to The Standard web site.  The results were much more powerful than I thought they would be.  Here's what's happening so far:

  1. Articles, blog entries, whatever all get tagged with freeform keywords via the popup tool added to my browser.  It takes about 15 seconds per article.
  2. Links are added to appropriate words within each article pointing to a new template on the site, passing a tag to that template.  For example, an article about Yahoo will now link the word "Yahoo" to a template on The Standard.
  3. The new template will display the collection of articles tagged with the keyword that was passed to it.  For example, the template may receive a request with the keyword "Yahoo".  The page will then pull all the articles with that keyword tag and display them on the page.
  4. Now, perhaps the most interesting part...the page will also display all the links that other users have stored under the same tag.  For example, the Yahoo page will show both the articles that The Standard has tagged and all the articles that the universe of users have tagged with the word "Yahoo".
  5. Finally, all the collections of articles have a corresponding RSS feed.  So, for example, if you wanted to know what The Standard or even the world was collecting in a bucket called "Yahoo", you could follow that conversation by subscribing to that RSS feed.
I've really just scratched the surface here.  There's a lot of power to this thing that I'm not sure I fully grasp yet.  One thing that I really like is that taxonomies so often become outdated the day you create them, not to mention a giant resource drain with all the meetings and revisions and implementation costs.  News evolves so quickly that you shouldn't be locked into a closed hierarchy.  So, this way I can intelligently tag and display all the rich content on this site, regardless of the type of content, as the stories we're covering evolve over time.  That gave me the ROI incentive to dive into this, and then I discovered deeper value which I'll share as new examples appear.

If this works, I'll detail a how-to on the steps that got me from there to here.  You'll likely notice some errors on some pages still.  Hopefully, those won't be difficult to fix. 

Jeffrey Veen offers a how-to for displaying content on your web site.  And Jon Udell offers a really cool screencast showing step-by-step how he uses with his own content.  He also starts to uncover the power of this technology in terms of the connections that start to happen between people.

The semantic web seems far-fetched on several levels, but I'm suddenly buying in to the idea that we can create it, that it might exist some day afterall.  More to come...



TrackBack URL:

No tags, but links to remember on topics of all sizes
Weblog:  clock -- watching time, the only true currency
Excerpt:  ETech and SXSW and IASummit all explored classification in many ways, according to various reports.
Posted:  Sat Mar 19 23:11:19 EST 2005
Lessons learned: The Industry Standard's still re-birth
Weblog:  Matt McAlister
Excerpt:  I handed over the reins of domain to my former colleagues this week.  I'm not sure what they plan to do with it now.  The brand may be worthless at this point, but no doubt there's a need for something similar to what that ...
Posted:  Mon Aug 01 00:50:24 EDT 2005
A new content distribution plan for online publishers
Weblog:  Matt McAlister
Excerpt:  Dick Costolo of Feedburner published his view of the future of RSS and where his company fits into it on the Feedburner blog.  Among several interesting ideas, he states a vision of the future based on the items within RSS feeds as opposed to t...
Posted:  Sat Nov 26 15:00:03 EST 2005
Top Stories from 2005
Weblog:  Matt McAlister
Excerpt:  The following posts here seemed to resonate with people or perhaps got some good link love last year:On Publishing Models...What is the future for Web sites in a world of RSS? (May 23)A Web 2.0 business model for publishers (September 8)How to present ...
Posted:  Mon Jan 09 08:00:05 EST 2006
Jon Udell, storytelling, and learning through imitation
Weblog:  Matt McAlister
Excerpt:  Jon Udell visited Yahoo! last week to share some of his thoughts on language, visualization, and storytelling, among other things that all connected together in the end despite a seemingly random journey through his recent thoughts.  One bit I rea...
Posted:  Tue Mar 21 12:44:58 EST 2006