This morning I was listening to Moira Gunn's interview with Alexis Gerard and Bob Goldstein, and the mention of lexicon got me thinking about the design of the tools to manage personal InfoSpace, and that led me to the thought that one's own Information Ecology continues to evolve... and that the suite of tools for managing personal InfoSpace needs to be sensitive to that evolution, and interlink

• a growing personal lexicon,
  which overlaps with
  an elaborating set of
• tags which may be shared with others in growing folksonomies,
  and
• interdigitates with the metadata that comes with images and sound files and other media types (texts, GIS layers, etc.), and
• facilitates searching and retrieval across all collections and media,
  and also
• retains pathways and traces of one's searches...
• and all of that needs to be in a Web service that follows the user from machine to machine... because many people use more than one machine, and most people get new machines every couple of years.

So the personal InfoSpace needs to persist (safe, always accessible) and must be flexible, designed to be ready to incorporate new media and new helper apps when they happen along, or when the user's interests and purposes morph in new directions.

In fact we all do this, via a gallimaufry of scattered and often not-interconnectable apps and devices (some of which can't be read on current hardware). We preside over cascades of old floppies, hard drives in obsolete machines in attics and basements, heaps of CD backups, bunches of DVD backups, stuff cached on servers with unpredictable half-lives, cluttered C: drives, collections of Web pages, perhaps several blogspaces... If there's any way to order such chaos and make way for barely-imaginable new devices, we must somehow conceive and build

• simple, flexible user interfaces,
• easily transportable containers and data structures,
  and
• ready means to explore and export and repurpose individual items.

It should be easy to tag/earmark an item, to explore sets of tagged items (with appropriate visualization), to augment any item's or (items') tags... some sort of semantic Web capabilities... ready connection via appropriate APIs of other services... sort and filter by date, by media type, etc... export to Web pages, blogs, RSS feeds... facilitate sharing where appropriate... and the look-and-feel should be easily personalized ...and wouldn't it be nice to have Uniform Resource Identifiers (URI) for every item? ...and some 'items' are simply pointers (tagged, perhaps, or otherwise augmented) to resources that could be anywhere...

What models are out there?

An "idea organizer" like OmniOutliner [Mac only]?
I've been pretty happy with many of the features of Onfolio ("Collect. Organize. Share.")

11 June
Every day (or so it seems) brings items that might fit into what I describe, or that nudge my ideas in new directions, or remind me that lots of other people who know much more than I do are working in these realms. This morning's Bloglines trawl brought It's about the Community Plumbing: The Social Aspects of Content Management Systems (Presentation submitted for Computers and Writing Online 2005, Co-authored by Charles Lowe and Dries Buytaert, May 2005) --about Blogs & CMSs | Collaboration & Social Networks | Drupal...

...and tagcloud.com beta, which I haven't figured out yet, but it uses the Yahoo Content Analysis API

By tapping into the Yahoo! web services, we have access to all kinds of content and search functionality. What used to be accomplished through dirty screen scraping hacks can now be done easier, quicker, and legitimately though a REST interface that can be implemented in most any programming language. The end effect is that people can mash web content into interesting ways that the original authors never intended. With the addition of cool tools like Greasemonkey, folks are bending the web into exciting new chimeras of hyperlinked goodness.