Here at education.au we have a number of experts in metadata management and information organisation. As you’d expect in that environment the rise of folksonomies has generated a great deal of discussion.
As part of a new project here at education.au I’ve been exploring how to combine the two into something I’ve named a taxonomy directed folksonomy (TDF). Basically we use a taxonomy (in this case ScOT) to suggest terms a user could in a tagging-type interface. It’s our hope that this interface will allow us to use the formally organized taxonomy at the back end to find related resources, while not intimidating users on the front end.
The folksonomy will also feed back into the taxonomy in that it will allow us to discover subject areas which aren’t covered well.
The screencast below shows how the UI looks at the moment (if you are reading this via a aggregator you may need to visit this post on the website for it to be displayed).
Click To Play
6 Comments
This is an exciting initiative, and I can’t wait to learn more about it. The concept is not at all new - some of us have endeavoured to achieve this in taxonomies —— but manually! By harnessing the power of taxonomy, and by building an adaptive hypermedia interface (you are doing this right?) to capture folksonomies, you can actually create a totally user-centric, yet robust taxonomy. VERY NICE!
Anyone interested in improving distributing cataloguing using controlled vocabularies has possibly come across this article already - this looks like a good forum to share it in:
http://www.dlib.org/dlib/january06/guy/01guy.html
I found this article a little reassuring during a time when I though I was considering the future of controlled vocabularies given the rise and rise of community book marking. I manage the Schools Online Thesaurus and the rise and rise of community bookmarking has certainly caught my attention.
What this article suggests, and Nick has demonstrated with Taxonomy Directed Folksonomies, is that distributed cataloguing need not assume distributed vocabularies. What I like about the demonstration of TDF is that the user does not know if the drop down list is generated by popular phrases or a vocabulary - of course over time the distinction should become blurred and the tags that are used should be a source of warrant for the vocabulary itself - more about that in another post I think.
Cheers,
Les.
Nick, how are you doing the predictive text type thing? Presumably client javascript working with a backend database? Is the script available anywhere, eg did you start from something openly available and then customise it? It sounds like a very useful thing to be able to do.
I’ve got something in mind where we would like someone writing their profile to choose ‘interests’ from a pre-existing list where possible, so as to encourage consistency of vocabulary, and add their own terms where necessary. Then people searching the profiles would also get the same unobtrusive help encouraging them to choose existing terms to search on.
Thanks for any pointers,
Ben Plumpton
Hi Ben,
I’m using the script.aculo.us autocomplete script to build the drop down thing. The backend is a Java servlet reading data from a database.
Your thoughts on how it would work are pretty much the way we are thinking, too. Look out for a paper on this, soon.
Nick, this is fascinating stuff. Thanks for sharing. We are gearing up to build a very similar application on our intranet in ColdFusion, with lots of AJAX and with hooks into a Google Search Appliance and a Montague Taxonomy system.
We’re mostly following del.icio.us for inspiration, though with a few additional bells and whistles, like thumbnailed screen shots, REST URLs, 5 star ratings and tinyurl.com -style redirects.
My biggest concern is whether we are organizing the data in the right way. I’ve been hearing a lot about RDF and the Semantic Web and I don’t really have a handle on it all. If you or any of your readers have any suggested readings, I’d appreciate it. And if you have time to field some questions, please drop me a line. Thanks again for sharing. Good stuff.
I think this is great, but there is still the underlying problem of needing a full taxonomy to feed the potential foksonomy. Isn´t that double mantainance work?
I mean, with enough resources (human and eco) then no problem but for a small budget and a need for specialisation it seems just too much. Although I wish this could be exportable!
Thanks for the news!
7 Trackbacks/Pingbacks
[…] Take this little gem for instance - Taxonomy Directed Folksonomies - just one tiny part of the coolness that was demonstrated. At today’s team meeting we decided to try and open up some dialogue with the world at large via blogs etc, probably in the ‘Show and Tell #3′ timeframe. […]
[…] Follow any comments here with the RSS feed for this post. Post a comment or leave a trackback: Trackback URL. « Humans will not survive theCentury […]
[…] My role has been to manage the Proof of Concept process. Some of the work has been blogged by Nick Lothian. It’s a process that I’ve really enjoyed and as well as been challenged. The challenging bits are more interesting so here’s some examples: […]
[…] As Nick points out in the comments I should link to an explanation of TDF. […]
[…] Patrick Lambe at Greenchameleon recently blogged a very positive reaction to a paper my colleague Sarah Hayman presented at the Developing and Improving Classification Schemes conference. His reaction was mostly to our concept of a Taxonomy directed Folksonomy, but he mentioned the idea of person-mediated serendipity. This is an area we agree is a vital concept, but we probably haven’t spoken about enough. (As an aside, this appears to be closely related to something Jon Udell refers to as “manufactured serendipity”, most recently in Data finds data, then people find people). One key concept our team has been exploring is the trokia of resources, tags and people. The idea is that users should be able to use any one of those starting points to find useful information in the other two. For instance, if you are interested in the “education” area you can go to the education tag and find both resources related to that area and people who do work in that field. […]
[…] The ability to group tags, e.g. del.icio.us’s “bundles”, provides one way for taxonomists to work with an underlying folksonomy. Another possible solution is a taxonomy-directed-folksonomy, which relies on the user interfaces to suggest tags from a formal taxonomy, but allows many users to use their own tags. […]
[…] Or, coming more “top-down” into either del.icio.us or LearningLinks with an existing authoritative taxonomy — if such exists in the learning area (note to self: dig into this more, the only thing even close to this that I am aware is the Learning Circuits glossary) — to guide the vocabulary and structure to more the ’standard’. This thought then takes me to this Taxonomy directed folksonomies blog post from December, which a colleague at Novell passed along: I’ve been exploring how to combine [taxonomy and folksonomy] into something I’ve named a taxonomy directed folksonomy (TDF). Basically we use a taxonomy (in this case ScOT) to suggest terms a user could in a tagging-type interface. I’s our hope that this interface will allow us to use the formally organized taxonomy at the back end to find related resources, while not intimidating users on the front end. […]
Post a Comment