The Death of Taxonomies! Is there a role for Semantic Technologies?

Image by justgrimes via Flickr

There is a very nice article by CMSwatch about how the practice of building taxonomies is dying - this discussion is more oriented towards content/document-centric world. It basically highlights few trends:

• Since social computing is being widely adapted by companies so there is no fixed way of categorizing things i.e days of single, heirarchical taxonomies are far behind. People will tag content in any way they want.
• You should be able to get to information the way you want, which may be different from your colleague's approach.
• Text mining and auto-tagging software is gradually improving, and extracted terms can be applied as metadata. Metadata needs to be very fluid - cloud like. See the fig. above
• Metadata architects should really understand the domain

I believe there are some solid arguments as I have seen this trend myself. I have only few things to add:

• What will be the future of existing large taxonomies?
• Many companies have already invested in more than one content management system like Sharepoint, Documentum or Opentext. There is a good opportunity for semantic web technologies, basically OWL and RDF, which can unify the taxonomies of different content management systems and provide a single data model to retrieve the content through.
• We should always be very clear about what the word "Metadata" means in any context. It means different things to different people and has been overused. A metadata in a document-centric world is entirely different from say a metadata in a word processing program.
• Semantic Web is really an umbrella language for metadata. More thought needs to be given to how this new trend of adding tags/metadata on the fly can be leveraged to add more value to the semantic web or linked data cloud.