Taxonomy Fairy Tales – Taxonomy Times

The YouTube video with Patrick Lambe a few months back generated some good feedback, including an invitation from Teresa Macgregor of the SLA Taxonomy Division to write it up for their newsletter. The following appeared in Taxonomy Times No. 6, April 2011.

It’s strange, but start talking to hard-bitten, seasoned executives about information in the enterprise and they automatically switch off their critical faculties. They’ll believe anything. Really. Like, information and how it is used in your organisation can be understood by a piece of software, out of the box. Like, you don’t need to actually understand your information environment in order to manage it. Like, the best people to ask about making your information generally accessible, are narrow subject matter specialists. Like, you can fix your information environment once, and it’ll stay fixed forever without paying any more attention to it. In this article we explore three fairy tales about taxonomies that executives seem particularly prone to believing:

  1. That you don’t need taxonomies if you get a good search engine;
  2. That taxonomies can look after themselves or can be delegated piecemeal to non-taxonomists;
  3. That the best people to advise on taxonomy development are subject matter experts.

Download Taxonomy Fairytales

This entry was posted in Articles & Papers and tagged . Bookmark the permalink.

7 Responses to Taxonomy Fairy Tales – Taxonomy Times

  1. Ian Fry says:

    Currently my semantic processing experiments are showing some interesting things about the tagging of action items in one of my research data sets. I am getting an interesting mismatch between the semantics of what is said, and the tags which have been applied from the taxonomy.
    We seem to have a mixture of
    – matches
    – errors
    – tags assigned where there was “no time” to spell out the associated implications of why the tag was used
    – tags not assigned because there were already a lot of tags on the item

    We could either be ending up using semantic searching and not using tags, or keeping tags with the semantic engine providing a QA to ellicit additional information.

  2. innotecture says:

    Ian – Can you into some more detail about your 4 outcomes? Can you give some examples? Why would you not use tags?

    • Ian Fry says:

      Still working on that categorisation with subject matter expert. I have given you very preliminary findings but will have this exercise nailed for my ACTKM talk.

      Automatic tagging is a “managed and repeatable process” which saves the time of a subject matter expert and may be more accurate / accurate enough – and therefore it is KM that can be “justified”

  3. Pingback: Taxonomy Fairy Tales – Taxonomy Times | Dam Foundation

  4. innotecture says:

    There’s definitely a place for automated tagging in lots of applications. Have you encountered any cases where a manual tag is concerned with the subtext/context of a piece of text than a semantic analysis would not pick up?

  5. Ian Fry says:

    Still working on that.
    What I have found is Tags being applied where there is no relevant associated text or meaning in the original text.
    In other words, the SME was using the Tags to highlight desired outcomes and relevances; but the actual text being passed on for action was devoid of any comment as to why or what to do.
    In this exersise we are tagging action items from the Lessons Learned with Treatments to be applied.
    This dichotomy will feature in my ACTM11 session.

  6. Pingback: Metadata 2011 – Australian Taxonomy Survey | Innotecture

Leave a Reply to Ian Fry Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s