First, a quick recap:
- Data growth is out of control — there’s too much stuff stored in too many places
- Data is growing way faster than we can use it, so more and more data is becoming less frequently used
- The combination of 1 & 2 undermine the efficiency of today’s data management tools
I also mentioned that hierarchical systems break down if too much information is added That leads me to another key point in our view of the universe: the traditional idea of structure needs to change.
The problem with permanent structure is that it doesn’t handle unforeseen exceptions. And the more data a system contains, the greater the odds are that there will be exceptions that need to be handled. The scramble to update systems to handle Y2K were a classic example of just how painful this can be.
We think it’s better to evolve structure into a transient user-driven thing that gets deployed at the time of consumption. It’s a mechanism to increase convenience at a particular point in time, rather than a foundation that a system is built upon. It turns out that meta-data (a.k.a. tags) can be a great tool to drive structure.
What we are building is a tag-driven system that unifies all your different data sources within a single interface. You could think of it like an RSS aggregator, but for any kind of data. Through a combination of smart and sneaky methods, we are automatically tagging data as a user organizes and interacts with it — no manual tagging is required. Those tags drive all kinds of useful functionality that helps you organize the data.
The end result is a personal library for all kinds of data that makes sense to you. Each individual library can be connected to other libraries, while maintaining your unique perspective on the world.
We’re getting ready to broaden our beta program. If you’d like to participate, use the "Contact" link on our home page to send us an email. All you’ll need is Firefox and a broadband internet connection.