I "met" Dorine Ruter when we both participated in the online Web 2.0 and Communities of Practice Online Conference. The tagging discussion included an action learning exercise where we shared a lot of resources about tagging as well as explored the discovery aspect of delicious.
Dorine is currently involved a pilot project for KM4DEV where a community of practice is collaborating on sharing delicious feeds. She recently posted about a demonstration she gave during a meeting with members of the "Dutch ‘e-collaboration’ group" about tagging and social-bookmarking.
In the comments, I prodded her to share some of her experience about using tagging as a tool to share resources in a community of practice context. She gave me some very rich reflections as a reply in the comments
Her key points:
She spoke with two other colleagues Allison Hewlitt at Bellanet and Peter Ballantyne of Eurforic who are also working with tagging in CoP/Organizational context and their comments resonated with her own observations:
The "how to" of CoP group tagging is an obstacle in getting to the "why" (exposing the intelligence of the group, thus making resources more findable.)
- What central tag to use to get an item noticed?
- Can anyone use this tag or do we want to create a special feed/account to provide filtered information? (BK: I wonder how the "network" feature or "private groups" feature of delicious influences this?)
- What tags to use to describe items, who decides this?
- What items to tag in the first place, do we need rules for this?
One of her points resonated with some other reflections I'm seeing on this side of the pond from early adopters in the nonprofit space in using tagging for resource sharing. Her point is this: Getting people motivated to tag content (easy tool, easy approach) and at the same time create good content as output (quality selection of sources, good descriptions and tags) is just very difficult.
She identifies some of the tension-points. I've translated these into some questions that nonprofits might discuss as part their experimentation with organizational tagging and perhaps leading to a tagging policy that works for the group:
- Who in our organization must be involved? Everyone or can it work within sub-group?
- How do we create an environment where people are eager to participate or do you focus on providing a high quality resource of information?
- What if people in our group are re-finding items tagged that others have tagged before?
- (Does that contribute to the willingness to participate or pollute the tag stream and create information overload?)
- What if items are tagged that are interesting, but not on the topic most of the group wants to focus on?
This correlates with some of Marc Sirkin's musings about his organization's experiment with tagging
Your tag or mine? Then there's the problem of naming your tags. Everyone has a slightly different way of thinking about things - I say Myeloma, you say Multiple Myeloma. Without a standard set of published, prompted tags, chaos could reign. Then again if you only allow pre-created tags, there's not flexibility for users to create their own world view within the tags. Trying to do both would result in "tag goo" (a term I literally just invented). (BK: Would it?)
Link Fishing
So let's say I have all these great tags... maybe hundreds of them. At that point, even a nicely formatted "tag cloud" would only serve up quantity, not quality of links - and from there I'd have to sift through perhaps tens or thousands of links in a much less efficient way than by just using Google from the start.
In the comments of this post, there is an interesting discussion thread that also points out the question, "How do you deal with information overload? Doesn't tagging create information overload? Why not just use search?"
The term information overload made me think of Marshall's fabulous interview with Robin Good on netsquared: "Mastering Information Overload." Good is advocating for a new type of knowledge worker - he calls them Newsmasters. These are human filters "who subscribe to the RSS feeds of a large number of sources, search queries, and other dynamic sources. The person uses a combination of machine automation and topic-specific expert knowledge. The most important information resulting from these filtered subscriptions is then delivered to end users, either by RSS, email, on a web site or by whatever delivery is appropriate. Central to the concept is that one person is responsible for curating information for others in the group or community." (The last part makes feel a little uncomfortable.)
So, how does the Newsmaster avoid information overload? Are these the early adopters of the Internet and whose brains have evolved along with the growth of electronic information and have superhuman prcocessing skills? (Not the early adopters suffering from Alzheimers) What's their secret?
Marshall notes in his article: "Newsmastering is all about acting as a curator of the huge new stream of information that is coming to us every day .... thus the decision often comes down to What can I afford not to read?" Ah, what Will Richardson did on Earth Day. Ah ha, the environmental metaphor calls to mind David Shenk's Data Smog: Surviving the Information Glut" and my favorite line in the book, "Give a hoot, don't data pollute."
Marshall also describes some simple technical approaches:
Newsmastering can be as simple as splicing multiple RSS feeds together into a mix with a tool like FeedDigest.com. The next step in sophistication is to filter your feeds and be specific in your sources. More complex topics require strategic choices in sources and filters.
Lest this appears as simple as just finding interesting things on the internet, Robin points out that not just anyone can be a good Newsmaster. The most important qualities to have are passion and competence in the particular field being covered. In order to select, edit and manage the real information gems in any sector, a Newsmaster must have a deep familiarity to recognize news and resources in their proper context.
So, no grand conclusions on any of this, but plenty of good questions for discussion and further experimentation and learning ....
Comments
You can follow this conversation by subscribing to the comment feed for this post.