Previous topic

Query language

Next topic

The about tag

Contents

Welcome

Users

When a new user (or application) is created in Fluidinfo, an object is made to hold information for the user. Fluidinfo stores some information about the user on this object.

The user is free to add more information to the object that represents them. Because Fluidinfo objects are not owned, other users or applications may do the same.

Social Networks

This makes it possible to build social networks inside Fluidinfo. As with all information in Fluidinfo, these can be built without asking permission and without needing to be anticipated. Because information can always be stored in its most natural location, interesting searches combining information on the same objects are immediately possible.

For example, suppose:

  • A developer writes an application that extracts follower information from the Twitter API. When run by Fluidinfo user Jack, it puts a tag called jack/i-follow onto the objects of other Fluidinfo users.

    It is immediately possible to do interesting queries. For example, has jack/i-follow except has lisa/i-follow to see people Jack follows but whom Lisa does not.

  • A second developer now writes a similar tool, perhaps pulling friend data out of Facebook, and putting USERNAME/fb-friend onto users’ objects in Fluidinfo. Searches across social networks are then possible. E.g., show me who I follow on Twitter but do not have a friend on Facebook, or show me who two of my friends know in common but whom I do not follow on Twitter.

Fluidinfo allows any number of examples along these lines. And because objects are not owned, you don’t need to wait for permission to try them out - just add your data and start searching. To see this in practice see our TechCrunch Disrupt presentation which illustrates the above using Tickery, We Met At, and Tunkrank.

And much more

Any information at all can be put onto other users’ objects. Think someone is a spammer and want to alert others? Tag them, and use the presence of your tag to avoid that user or their content when querying. Want to indicate people you trust? Tag them, and use the tag in searches.

Not happy with Twitter’s Suggested Users List? Why not make your own and do searches that combine yours with other people’s recommendations? Want to put ratings onto tweets, or put your own measure of interestingness onto users? Just go ahead and do it. Then search on it, combine things, and mash things up to your heart’s content. See how to do this via our article Putting metadata onto tweets with FluidDB

Similarly, namespaces and tags can also be tagged and searched on. In this way Fluidinfo allows for deep personalization and for the evolution of reputation and trust.