What are the personal benefits of tagging? (Part 1) May 15, 2006
Posted by Steve in : enterprise, tagging , trackbackAs I mentioned in my previous article about social bookmarking, we’d like to ensure that the social aspects of our application achieve significant enterprise adoption, so we’re attempting to follow a lesson learned from del.icio.us: that personal value precedes network value. I’m hoping that others might benefit from our thoughts on value, or contribute new ideas, so I’m speaking them out loud here…
I don’t think it’s possible to state the value of tagging to a uniform “application user”, rather, it’s important to consider a full spectrum of participation roles:
(This image is from that excellent article, by Ross Mayfield)
To simplify things a little for this article and the next, I’m going to approach tagging from two perspectives – a consumer’s and a publisher’s – which cover the spectrum pretty well. In each case, I’ll try to isolate benefits that are immediate and specific to each particular role, rather than the more general “social” benefits everyone gets from sharing.
Benefits to consumers
So, first, what great personal value can “read-only” consumers get by tagging their bookmarks?
The most obvious reason for a bookmarker to tag is so that he can categorize and organize collections of bookmarks. Anyone who’s saved more than a handful of links can see the benefit, and quickly picks up the metaphor: tagging is like putting a bunch of items into a shared folder. Note however, that new enterprise taggers may take a bit of time to get used to the fact that items can be in more than one folder, especially if some of their folders (public tags) can contain items that they didn’t put there themselves! (In an upcoming article, I’ll talk about some specific challenges that we’ve faced over about two years of enterprise “tagging”.)
Several tag-as-folder patterns have emerged to store common collections: “save for later” tags, shopping lists, etc. These patterns will be just as applicable to enterprise users, and are great examples of user needs that haven’t yet been met by other applications, but can be easily built on top of the generic folder-ness of tags.
Many people also use tags as a way to summarize bookmark contents. Familiar with this usage on sites like Flickr, as well as any number of legacy apps with keywords or synopsis fields, they adopt key document phrases as tags. The end result is that the tag line of such entries often reads like a mini-abstract of the article. Tom Coates of Yahoo suggests that this method of tagging may be becoming more common, but I’m not sure that this practice is actually that useful. I see little return on the investment of extracting keywords, when search engines do a far better job of it. On the other hand, a quick analysis of my own tags suggests that I originally tended to tag this way, even if I don’t find benefit from it!
Another use that I’ve observed among more organized bookmarkers is the use of tags to store metadata where no specialized fields have been provided. I’ve seen tags that encapsulate authorship and/or location, company names mentioned in the article, referring connections (“via: whoever”), etc. Again, I haven’t yet found much value in doing this. It feels to me like exactly the kind of usage that only punishes you for not doing it 100% of the time, especially when a search index has already ferreted out most of those pieces of information. Perhaps not coincidentally, though, the heavy users of metadata tags tend to be librarians. They definitely know more about organization and recall than I do – maybe I’m just missing something?
Once a particular installation has built up enough of a bookmarking profile, social effects that were previously a long-term, group benefit may create immediate, personal benefits to new users. For instance, users adding bookmarks to a well-populated system can immediately receive similar item recommendations. As with folder-like usage of tags, this is a personal benefit that increases over the user’s lifetime of usage, as well as with the maturity of the system itself. This is also a distinct area of improvement for most social bookmarking systems, where the extent of “relatedness” is sharing a common tag. Better group-similarity analysis will make a huge difference in the next generation of social systems (more about this in a later post…)
Finally, I’ve seen some very interesting analysis of tag clouds that go far beyond the original uses of tagging. For instance, Jon Galloway has started using his metadata cloud as a way to discover trends in his own attention that he might not have otherwise noticed. Juice analytics does something similar. Have you come across any similarly ingenious applications that directly benefit the tagging consumer? Let me know!
One thing that the most useful of these reasons all have in common is that they allow the user to express tags using personal vocabulary. It takes a lot of work to settle upon a public vocabulary for any particular topic. As evidenced by endlessly recurring “what tag should we use for this?” conversations, the right terms might not even exist at the time of tagging! A purely selfish tagger has very little incentive to use common, group-accepted terms over words he uses every day. So one thing to assume while developing an enterprise app is that users do not want to be taught how to tag, at least not before they’ve gotten enough personal benefit to start accepting tagging.
Benefits to publishers
The next post in this series tackles the value of tagging to a publisher… Meanwhile, do you think I’ve missed any interesting uses of tags?
related: tagging benefits
related: tagging visualization
Tags:bookmarks enterprise social tagging
Comments»
Steve,
Thanks for your article on tag-based classification. Want to let you know that I put these ideas to life at http://talkinghub.com/ which I welcome you to take a look and leave some comments.
Denis Krukovsky
http://talkinghub.com/