Community, Database Style

In session after session at the recent InfoCommerce Conference, the word "community" was mentioned repeatedly. Whether it was in the context of a product feature or an entire business model, community seems to be hot.

I've generally been very dismissive of the whole concept of online community. While it remains a popular buzzword, its meaning is tough to pin down. If we agree for the purposes of this discussion that community is some form of online system and venue for users to interact with each other, then community has been with us since the earliest days of the Internet. Indeed, community in the form of chat rooms fueled the early growth of services like AOL. During the dot com era, it was seemingly de rigueur for every new publisher site to have a community feature. Why? They were cheap, ran themselves, created free new content and were believed to make a site more sticky. All good things, except for one small problem: they rarely if ever worked successfully.

Since then, online communities have taken many forms, but very few have been databases. Yet community-built databases have a good track record. Look at the current success of Wikipedia, a community attempt to build a free online encyclopedia, or look at the Internet Movie Database (which was a volunteer, community effort until acquired by Amazon). A case can be made that there is greater willingness to contribute to building a truly useful reference than there is to slinging questions, comments and insults back and forth. Certainly, the issue of externality exists with a community database (though clever incentive programs tying access to contributions might perhaps address this), but a community database seems to have a lot less baggage than other types of online communities. Time is less of an issue since a growing base of reference data is being constructed. Anonymity is less of an issue since the content is rarely the least bit controversial. Noise gets filtered out through the nature of the community and its structured input.

I certainly agree that a community database is less interactive than a community chat room. But though it sounds counterintuitive, it is looking like you can create a more sustainable community by limiting and structuring interaction. And who owns all the valuable content that users are contributing? Whoever hosts and structures the interaction, and nobody knows structured data better than data publishers. Opportunity knocks.