Taxonomies and Folksonomies

Knowledgebird word cloud

I’ve observed a discussion on the twitters lately dissing hierarchical categorisation in favour of tagging. The conversation was around incident management, but I think it’s worthy of expanding on in a broader knowledge management sense. It’s well-known that I like it both ways, but this rage against the machine* is misguided and potentially ineffectual. (* Where machine represents formal taxonomy.)

First of all, what is taxonomy and folksonomy?

Folksonomy is a way of describing crowd-driven categorisation of information. This flat, loosely-structured form of classification via tags became more widely used around the time of social bookmarking sites like delicious becoming popular.

Taxonomy, conversely, is a formal, hierarchical method of classification. It recognises that cheddar is a sub-class of cheese.

The case for taxonomy

Folksonomies are surely charming. They’re compelling and serendipitous; they capture somewhat of the collective’s personality. As communities of users develop their own naming conventions, folksonomies capture symbolism from the meme du jour, variations and typos, too. Depending on the reason for retaining a knowledge exchange, maybe that’s just fine. Go back to the strategy and consider the goals of retaining and maintaining whatever kind of knowledge you might be dealing with. If all you need is some run-of-the-mill search engine logic, then it’s all good. And, if there’s some rules-based automation available, tags can add some powerful efficiencies.

One argument might be that the artificial intelligence of search engines powerful and over time, with further advancement perhaps, search functionality will be able to connect the dots of folksonomies and draw accurate conclusions about the information you want to find. But in mimicking human thought and our fuzzy logic, search will only ever provide what you probably want to know—based on like terms and popular votes. Like I said before, maybe that’s good enough, but it just isn’t scalable.

Remember when we used to have to defrag our computers? You know… before we all got Macs? Computers like order. It helps them find the files quicker. When the Singularity comes we might still be charming, folksy humanoids but the machines will have us all completely catalogued and ordered, and functioning in exactly the way they expect us to and in the most efficient way possible.

Taxonomies might feel stuffy and dictatorial, but it makes for efficient searching, sorting and reporting. The common language borne from taxonomy is a basis for understanding and collaboration, and they CAN be fluid if reviewed and communicated regularly.

And like I’ve said before, they can live in harmony together. WordPress is just one system that allows for both hierarchical taxonomy, in the form of post categories, and less-structured folksonomy, in the form of tags.

Vive la différence!