Taxonomy software

From WhyNotWiki

Taxonomy  edit   (Category  edit)

[edit] Open Taxonomy

	"Collaborative Content Searching and Indexing"
Homepage:	http://opentaxonomy.sourceforge.net/
Project/Development:	http://opentaxonomy.sourceforge.net/
As listed in other directories:	http://swik.net/open-taxonomy
License:	[[:LGPL|LGPL]]

Open Taxonomy - A Collaborative Web Directory (http://opentaxonomy.sourceforge.net/). Retrieved on 2007-03-09 17:16.

[edit] What is Open Taxonomy?

Open Taxonomy is a web based collaborative web directory system. It allows users to navigate a tag based taxonomy that has been created by them. Users can share their categories, or tags with others, then vote on and rank each other's tags. Users can search the indexed web content for keywords in specified sub-trees of the taxonomy, which greatly increases the content's findability. The system is built using Ruby on Rails to provide a rich application experience to the user in a web browser.

[edit] Ever spent more than an hour searching for something on the web?

It's too hard to find what you're looking for, either on the web, or in an enterprise environment. Keyword searching is ambiguous. Search for "soap", and you'll get results about detergents, daytime TV, and an internet protocol.

Social Networking sites like del.icio.us and FlikR.com allow users to share tags they have placed on concepts, and link to them. This helps, but the tags form a flat vocabulary. There's no structure to them. It's still difficult to locate things.

Web directories organize everything into categories, and sub-categories, forming a taxonomy of all information. They are relatively difficult to change, and as new content is created and indexed, it is a constant battle for a static taxonomy to keep from getting stale.

Open Taxonomy allows end users to search the taxonomy for a concept, and if they can't find it, they can fix the taxonomy themselves, so that the next person can drill down directly to it. They do this by tagging the concepts, and tagging the tags with a parent category, eventually merging the tag structure into the "official" taxonomy.

Ken Cooley. a talk on the Open Taxonomy project at the June 10 2006 BarCampHouston unconference (http://barcamp.org/f/OpenTaxonomyBarCampHouston.ppt). Retrieved on 2007-03-09 17:16.

• allows navigating a directory of categories
• Users can create their own category tags, and link content to them.
• They can tag existing tags, adding additional categories anywhere in the bottom half of the taxonomy.
• Other users vote on these tags to make them part of the official taxonomy.
• Navigation
  • Tree of categories in left window.
  • Icons denote type, sub-category, synonym, part-of, etc.
  • Current location (in the tree) listed at the top.
  • Indexed content in the center
• Adding or changing categories
  • Tag the tags (one tag is a sub-category of another).
  • A few other relationships supported.
    • Part-of
    • Attribute-of
    • Related-to
  • Parent tags could be constrained to a level in the tree. [?]
• Incentives to use it (self interest)
  • If it can’t be found, fix the taxonomy, so it can be found next time.
  • Others will make other things easier for you to find (the Wikipedia golden rule).
• What happens if too many different ways of organizing the same info get elected?
  • The administrator impeaches (un-elects) select categories to streamline the tree.
  • Old links would still work
  • The original contributors would still see them.
• Control the election threshold as a function of the # of existing alternate paths to the same content. [?]
• Too Complicated?
  • In the simplest case, it degenerates into just another simple tagging site. The structure isn’t required.
  • If someone wants to add structure to unstructured tags it won’t make it any more difficult to use.
• Enterprise Considerations
  • Threshold for getting a category elected could be reduced to a couple of votes at the leaf node level.
  • Could be spliced onto a fixed corporate taxonomy.
  • Site-dependent meta-data templates for all content indexed (date, project, version, department, etc.).
• Future additions
  • SOA service
  • Import/export
    • OWL (Semantic Web Ontology)
    • SKOS (W3C Thesaurus)
    • XBRL (Extended Business Reporting Language)
  • Use Corporate User management systems.
  • Synchronize with Corporate Taxonomies
• Merge [these efforts/processes into one]:
  • Tagging
  • Grass roots content indexing
  • Web Directories