Yahoo’s Tom Coates coined the phrase “a web of data,” using it to describe the current state of
the Internet and the transition from many websites serving as isolated repositories of data to a much
more interconnected model, which allows data to be shared and manipulated throughout the online
Imagine that you are the owner of a database full of restaurant addresses. You can display them in a
list, sort them, filter them, and allow users to search through them; however, you cannot show your
users where they are located on a map because you don’t have any maps stored, and even if you did,
you don’t have the skills, time, or desire to create the logic that takes an address and pinpoints its
exact location on a map.
Now say there’s a second company that owns a wealth of mapping data that can be searched by
address. Separately, you have two moderately useful websites; a user can search one, but in order to
make sense of the data they find, they have to visit the second website.
Obviously, this is not the best experience from the user’s perspective, because they need to visit two
websites in order to find the information they need. When information is arranged in isolated silos,
the user experience usually suffers.
If you examine the same scenario but with open data sources, you will see new possibilities for innovation.
For instance, if the owner of the address data had access to mapping functionality, then they
would be able to provide customers with customized maps for each address in the database.
The user would benefit because the information they are looking for is more readily available, without
requiring visits to multiple websites. The benefits can work both ways, though; the owner of the mapping
data might want to allow users to search for restaurants. A search term could be cross-referenced
with restaurant names from the first database, and if there are any matches, then they could be displayed
on the map.
And that is just the beginning.
Many developers have taken this much further, developing innovative new web applications by mixing
different sets of data and functionality to create mashups.