Folksaurus is a user-driven controlled vocabulary with an API.
You're probably familiar with tags, the terms used in web applications to describe resources, be they photos, blog posts, or whatever. These terms are completely free-form, entered in by end-users who just type in whatever terms they think other users are most likely to type into a search field. This practice works surprisingly well, but it does have some drawbacks.
First, the free-form nature means that people will use many different terms to describe the same thing. On the flip side of that, people will also use the same term to describe many different things. The result is that when you search on a tag, you'll get some irrelevant results, plus you'll miss out on relevant ones.
Second, tags lack relationships. While some web applications will analyze the usage of tags and conclude that two tags are related because they are often both used to describe the same resource, the app does not know how the terms are related. It may associate "cows" with "barns", but it does not know that they are both things found on a farm. It may associate "dogs" with "dalmatians", but it does not know that a dalmatian is a type of dog. Having data on the relationships between terms greatly improves the ability to browse.
The answer to these issues in the library world is to use a controlled vocabulary, a set of terms and term relationships compiled by information professionals. These have their drawbacks as well.
First, they are designed for, not just by, information professionals. Getting untrained users to adhere to their correct use is not likely.
Second, they are limited. If a term does not yet exist in a vocabulary, it must be added. In order for it to be added, the aforementioned information professionals must review and agree upon the new addition.
Third, the terms agreed upon by the compilers of the vocabulary may not always correspond to what most people would consider to be the preferred term. Somewhat famously, at least in librarian circles, the preferred term in the Library of Congress Subject Headings for "cooking" was "cookery" (until it was finally corrected in 2010).
The goal of Folksaurus is to combine these two systems, resulting in a service that has all of the benefits of a folksonomy and a thesaurus, but none of the drawbacks. It consists of two parts: the web service and the website.
Anyone may visit the website and make changes to the thesaurus itself. A reputation system is in place to prevent users from making certain major changes, such as changing the name of a term, without first making a number of more minor changes which are rated positively by other users.
In addition to requesting term details from Folksaurus, an application may submit a new term which is added to the vocabulary as an "unsorted" term. Users of the website may then edit the record for an unsorted term and specify whether it refers to a concept not yet represented in the thesaurus, whether it is another word for an existing concept, or whether it is ambiguous, referring to one of any number of different concepts. Users may also categorize the newly submitted term by specifying "broader" and "narrower" terms. Once such relationships are defined, the term is no longer "unsorted". The next time a Folksaurus-powered app sends a request for this particular term, this new information will be returned to the web app, which will use it to create a more precise searching, browsing, and tagging experience for the end-user.
That's the idea at least. For now, the project is still experimental. If you'd like to help with the experiment, you can do so in a number of ways.
- Check out the site and offer feedback, suggestions, or bug reports via the contact page.
- Use the service for your own application and offer feedback on your experience.
- Write a library or plugin.
- Spread the word.