On distributed tools and mashups

Stephen comments on my recent As We May Interact?:

Tools like Flickr, Friendster and Technorati each try to become, if you will, a destination for people, to aggregate as many users as they can. We need to focus less on these big centres and more on how even unpopular tools can be mashed up and aggregated. There needs to be, if you will, a long tail of Web 2.0 tools – but nobody knows how to do that yet.

I think we may know how to mashup “even unpopular tools,” depending on what one means by mashup. By mashup, I mean utilizing a wide range of individual tools (like flickr, delicious, technorati, etc.) and aggregating the results of those uses into a collection of data that I can do new things with. If this is what we mean by mashup, I think that RSS and our imaginations give us most of the answers we need.

I also think we have to be careful in how we approach the decentralization issue in this next generation of tools. Let’s not forget what Ritter taught us: if we fully distribute data and go searching around for it whenever we need it, it doesn’t take that many users before the query traffic eats up all available bandwidth – never mind transferring any of the data you actually were looking for. We need an approach that distributes specialized data-creation functionality across individual tools, and centrally aggregates the results of those activities for convenient, synergistic use via a variety of other tools.