Are the terms ontology, taxonomy, and folksonomy interchangeable?

December 18, 2005 5:39 PM

Are the terms ontology, taxonomy, and folksonomy interchangeable?

I started a new job and had a first meeting on Friday with some other departments. My position is as a writer, but I have to work with the Web people and have input into the "architecture" if you will. That being said, are the terms ontology, taxonomy, and folksonomy interchangeable? I know they talk about the same realm of stuff, but I didn't think they were the same thing. The terms were used in regards to a redesign and restructuring of a huge site.

posted by nramsey to computers & internet (17 comments total)

Ontology:

n : the metaphysical study of the nature of being and existence

Taxonomy:

1. The classification of organisms in an ordered system that indicates natural relationships.
2. The science, laws, or principles of classification; systematics.
3. Division into ordered groups or categories: “Scholars have been laboring to develop a taxonomy of young killers” (Aric Press).

I'd say they are not interchangeable. Unless there is some new ad agency speak that the dictionary hasn't heard of yet.
posted by spicynuts at 5:57 PM on December 18, 2005

I suppose people often use ontology and taxonomy interchangeably (incorrectly so, but harmlessly incorrect, usually), but you wouldn't throw folksonomy into the same bag, since it's very distinct from both of them.

Take this with a grain of salt, but here's my best try:

A taxonomy is just a classification of things. They are usually hierarchical. They do two things: give exact names for everything you're dealing with (your 'domain') and show which things are parts of other things (sometimes called parent-child relationships, sometimes called broader-narrower).

An ontology is like a taxonomy in that it is going to contain all the entities in your domain (for one reason or another--probably its roots in philosophy--people often seem to use the term "universe" when talking about the domain of an ontology), and show the relationships they have to each other. However, it does more: it has strict, formal rules (a "grammar") about those relationships that let you make meaningful, precise statements about your entities/relationships.

When I think of an ontology, I think of putting the universe in a bottle. It's a very ambitious thing to do. If you have a proper ontology worked out, it means you know everything about everything. In general, the more useful an ontology would be, the closer to impossible it's going to be to make it.

Folksonomy is a new term -- I mean, literally, it came about in the last couple of years. It's essentially what you see on del.icio.us or flickr. It's in many ways the exact opposite of a taxonomy in that:

a. Folksonomies are flat (that is, they have no hierarchy, and show no parent-child relationships)
b. Folksonomies are completely uncontrolled (part of making a taxonomy is deciding what the names of your entities are, but in a folksonomy, there can be a thousand different words for the same thing)

Any relationships you see in a folksonomy have to be derived mathematically (statistical clustering).

However, a folksonomy is like a taxonomy in that they share the same purpose: classification. A lot of debate is going on right now about which is better, though of course the answer is that you use them for different purposes. It's nice to have both.

By the way, last year, my Classification Theory professor (a very smart guy) expressed similar uncertainty about the practical distinction between taxonomies and ontologies, so we aren't alone in our confusion.
posted by Hildago at 6:14 PM on December 18, 2005

[fixed spicynuts' definitions]
posted by jessamyn at 6:22 PM on December 18, 2005

They are interchangeable in that they are all constituents of the same medley of jargon people sprinkle liberally over their sentences to sound important, but that's where--more or else--the similarities end. There was a good discussion of ontology in this AskMeFi thread.
posted by ori at 6:29 PM on December 18, 2005

A good rule of thumb? If you're not doing set theory, symbolic logic, philosophy or mathematics etc, don't use ontology.
posted by miniape at 6:33 PM on December 18, 2005

What miniape said: ontology is only properly used, IMNSHO, if you're doing philosophy.

A Taxonomy is an organizational schema for objects or ideas: the Linnean classification system for species, the Dewey decimal system, Library of Congress sytem, etc.

A Folksonomy is not an organizational schema at all...instead, it's the psuedo-organization that grows up around the social tagging of objects, a la del.icio.us or flickr.
posted by griffey at 7:08 PM on December 18, 2005

Miniape, griffey, thanks for the short answers. And Hildago, thanks for the extensive explanation, now I don't feel like I have to catch up on unnecessary jargon. It was pretty much what I already knew but there was a moment of new job jitters that made me doubt myself.
posted by nramsey at 7:36 PM on December 18, 2005

I was going to try to write something long, but Hildago did better than I did in our Classification Theory class, so I'll defer to him.
However, there are some examples of taxonomies that are much more simple than Library of Congress Subject Headings (LCSH) or Dewey: browse the Seattle City Clerk's Thesaurus, a very small and specific taxonomy, for example. It's pretty simple and has few enough terms that you can probably figure most of it out just by looking at it.
Also, taxonomies are often invisible in non-library situations. For example, microsoft.com, corbis.com, and gettyimages.com, among many others, all use taxonomies in their search engines, but you wouldn't necessarily know it from doing a search. Folksonomies are, by nature, extremely visible.
posted by librarina at 7:41 PM on December 18, 2005

Here's a link to SUMO, since I'm coming up short links to examples of ontologies. It's an OK example, but it's kind of hard to get information out of that site.

You do sometimes see the word ontology used in web development: it pops up a lot in discussions about the semantic web, for instance. Like folksonomy (in my opinion) it's kind of a web buzzword, but there could be a very good reason for your co-workers to be throwing it around, depending on what they're building.
posted by Hildago at 9:02 PM on December 18, 2005

Ontology in the context of information architecture, according to BoxesAndArrows.com: "Network of concepts and relationships between them. An example could be a ontology of wines, with relationships between each vintage to indicate its type, producing, taste qualities, food it is best served with, etc. Because all the concepts are discrete and important, and because the all the relationships between them are semantically structured, we can infer a lot of information from an Ontology that one just can’t from crawling the links on a series of web pages. (The question of how to take the spaghetti diagram of an ontology and turn it into a comprehensible set of linked web pages is addressed later.) What makes a formal ontology more robust than a thesaurus or faceted classification is rich semantic relationships, semantic restrictions on relations, range, domain, cardinality, logical sets, inverse relationships, etc."

They have a number of articles that might be helpful in relating to people in this business, such as Building a Metadata-Based Website referenced here, and Developing and Creatively Leveraging Hierarchical Metadata and Taxonomy.

Also, here is a good paper on ontology using wine as the example domain.
posted by Tubes at 10:17 PM on December 18, 2005

Why on earth would anyone take a harmless, obscure word like ontology with a clear meaning necessary to a certain limited group of people and decide to toss it into the jargon blender, turning it into yet another vague, near-meaningless word sort-of-synonymous with a bunch of other vague, near-meaningless words? As a hardshell descriptivist, I realize the cat's out of the bag and it's hopeless trying to retrieve the word, but it's a goddam shame, and I just don't understand the motivation. Why would anyone want to talk about the "ontology of wine"? English is a rich language; there's no need to pull that crap.
posted by languagehat at 5:43 AM on December 19, 2005

Mr Hat--although "ontology" as used in computer science seemingly strays pretty far from its original use in philosophy, its use there is actually well-defined. It's a way of making structured SVO assertions about the world, such as "Alice knows Bob" or "Riesling originates in Germany." This is what RDF is used for (its use as a syndication format is almost incidental). These assertions, in computer-ontology-land, are called "triples," but there are other, more complicated ontological grammars that allow for "n-tuples" so that you can have conditional truth values on your assertions (technically, I think there's a way to do that in RDF too, but I'm getting in over my head).
posted by adamrice at 6:32 AM on December 19, 2005

languagehat: because it sounds impressive, and people are insecure.
posted by matildaben at 6:33 AM on December 19, 2005

Soergel wrote a great paper difference between a taxonomy and an ontology. Essentially, a standard taxonomy/thesaurus/controlled vocabulary has only three relationships, "broader term", "narrower term," and "related term." An ontology has a mess of sematic relationships between terms.

In a taxonomy, for example, you could have:

Dogs
      Related term: Housepets
      Narrower term: puppies, border collies, Fido
      Broader term: Housepets

An ontology could add a-whole-nother layer of meaning on top of this. You could have:

Dogs
     LivesIn: House
     Chases: Cats
     CrapsOn: Carpeting

Also, these are not buzzwords (with the possible exception of "folksonomy."), classification and knowledge organization have been going on for a long, long time and there's a mess of scholarship and theory behind it. If librarianship seems buzzwordy now, it's because the Web has made enough people aware of the problem of classifying and finding information that it seems like our field has sprung up overnight.
posted by stet at 9:45 AM on December 19, 2005 [1 favorite]

Can we get the man who coined the term folksonomy, Tom Vander Wal, in here to comment?
posted by joeclark at 10:53 AM on December 19, 2005

Stet wins.
posted by Hildago at 12:52 PM on December 19, 2005

Naw, Hildago, it's all you. I just had a citation laying around from an independent study with our classification theory professor.
posted by stet at 3:38 PM on December 19, 2005