So now, what about RDF (Resource Description Framework)? (If you need to catch up, read my post on Web 2.0.) And what do I mean by ‘everything and everywhere’? The everything refers to interconnected databases that use RDF to relate elements between databases by using their metadata. To use a simple example:
Metadata about Chicago could be ‘city’, could be ‘United States’, could be ‘Illinois’ and many more. You have seen this before–it’s otherwise known as ‘tags’. Tags offer the benefit of flexibility, but they do not offer consistency from person to person. For the ‘everything’ part of this equation to work, you need a flexible and consistent way to search the data.
Right now, when you perform a search, you are searching the presentation of the data (the HTML code that composes the data in a human readable format). Unless you have access to the database that holds the data, you are not searching the raw data. In part, you can’t do that because you don’t know how the data is organized.
Visualize this: the last scene in “Raiders of the Lost Ark”, when they wheel the ark into a HUGE warehouse, with no signs or labels, with thousands of similar crates. The reason this scene is so funny is that everyone realizes that this is a MUCH safer place to hide the ark, because nobody can find it without any signs. (If you have never seen it, here’s a clip–but I strongly recommend watching the full movie. Very fun!)
The reason the ark is safe is the same reason you can’t search databases directly: you need an understanding of how the ‘warehouse’ is set up. This is what RDF is going to do for us, not just in a single warehouse but across all the warehouses in the world. It is going to create a simple framework using the metadata to organize data. This means that I don’t have to know how your database is organized to find the data I am interested in. So in the example of “Chicago”, all I have to know that one of the RDF elements is “city”. If your database and 10 other databases are using RDF, I can find references to “Chicago” in all 11 databases.
Instead of being hampered by invisible structures (how the data has been organized), I now can search and combine data in brand new ways to me (and maybe to others as well). This is called ‘linked data’. As an example, I can combine data from veterinarians with longevity data and probate records to prove (or disprove) that pet owners live longer and leave more assets than those who don’t have a pet. This is just an example, of course, but certainly not a ridiculous one. Finding unexpected correlations such as this (hypothetical) one are exactly what the ‘everything, everywhere’ features of Web 3.0 will bring.
Now to the key question–why should you care? You are probably not a programmer or computer scientist. But just imagine the possibilities.
What if you are trying to decide where to buy a house, and you are interested in knowing as much as possible. Right now, you are limited to searching sites for information (and connections) that the website owner has deemed relevant. But what if you have some linked data that is unique to you? Continuing with our example, what if you want to search for a house in certain price range, within 2 miles of a dog park and a well reputed veterinarian? Each search will be separate, and you will need to amalgamate the results yourself. With the eventual adoption of RDF, you will be able to run searches that combine different, disparate data in combinations that are unique to you. This will help you make more informed decisions, both small and large.