Research e-Labs

Why a community circle - sounds kind of old fashioned.  I’m thinking of the tendency for successful online communities to circle through online and real physical spaces and back again.  What technologies should we use to to make the online space richer and “more real”?

Thoughts on Webstock 2009

One of the interesting things about the most successful online communities seems to be the desire of participants to do stuff together in the real world.  Members of some groups at close physical locations meet together for events e.g. Flickr communities, others make physical things together like publishing a real paper magazine.  Despite the many advantages of online communications, most of us still seem to crave direct physical and social experience to make it “more real”.

So how does this affect the way technology can be used successfully.  Perhaps it is around the level of engagement.  If you have dedicated emotionally committed members of a community they will want to make it “more real”.  So the idea of a community “moving online” is really just part of a natural life cycle, not an end point.

The best definition I heard was this:  “Web communities happen when people are given the tools to use their voice in a public and immediate way, forming intimate relationships over time” from Derek Powazek.

Technology

Technology provides the tools but can’t make the community - it is only part of the mix.  So let’s assume for now that we have a real community.  We’ve dealt with the privacy policies and all the other stuff.  What tools do we need to provide to make this work beyond the obvious things that come out of the box these days (forums, blogs, wikis etc.).  Here are some ideas that surfaced during and after a workshop run by Derek Powazek and Heather Champ.

It seems to me that user tools fall into two main categories - those driven entirely by the user and those which use automation to give the user enhanced choices.

The first category includes tools like star ratings, user reviews and comments, creating friend networks as so on.  The second category starts to make things interesting by mining site data to expose the user to extra information.  The Amazon links that show “people who bought this also bought … ” are an obvious example using analysis of order data.  Check out this interview with Toby Segaran and his book if you want to dig into the statistics (not too complex) and real programing examples.

These techniques can also be used on sites that aren’t trying to sell stuff.  For example you could provide search aids - “users who searched for this type of information also looked at …” etc.  Or you could let individuals flag issues and use algorithms to escalate persistent stuff up to admins and respond back to users.  Even the biggest sites can be made more personal this way and encourage more engagement.

There are of course problems when you’re trying to provide a “personal” service when it’s really based on statistics.  People often have wide tastes which change over time.  I find my Amazon recommendations are pretty mixed and it’s sometimes hard to see how something got picked.  Constant experimentation and improvement of the algorithms seems to be the way forward.  From a technical point of view developers need to set expectations that the first version is just that - a step on the way.

Government

Are recommendation techniques being applied on government sites?  If they are, where are the best (and worst) examples?  If not, how should government be applying the technology?

(179 votes, average: 2.8 out of 5)

This entry was posted on Wednesday, March 4th, 2009 at 4:58 pm under Standards & Compliance. You can leave a response, or trackback from your site.