I’ve been blogging a bit over on Nodalities about “stuff being connected”. The idea being basically: everyone is constantly creating data—all the bits of information that can be used in abstract. These tiny bits of information are constantly being generated by every process we undertake, from the obvious like online banking to the more obscure like driving to work (your odometer tells you how many miles you’ve gone, your on-board computer may store info about your car’s status, your satnav knows where you’re going and been, your mobile phone may know this too, the garage knows when your last service was… this list can go on and on). These data are more powerful when automated by software, and they become exponentially more useful when they are connected with other data. For example, the knowledge that £50 pounds left your account isn’t particularly helpful without a connection to that little bit of data which tells you the date of the transaction.
But why are some data more obscure—why don’t we even think about using some of them?
It may be simply because they’re not immediately useful to us, yet. We can, right now, log in to our banks and have a look at our accounts. We can shuffle and access and compare and analyse because this information is being presented to us in an easily-managed and understandable way. We have access to the raw data, and most of us have some basic understanding of why these data are important. I wouldn’t be surprised if readers of this blog have a spreadsheet or two with financial calculations on it, or use quicken with their balance info. We all know how important calendar events, emails, address book contacts, and bank balances are, and we have various systems to deal with them.
But, what do we DO with all the data we don’t currently access routinely? Well, this is where those connections come in. We can connect data together using some sort of framework, or abstract construct like a database. However, this database will need to be connected to another database (or exported to an existing one) in order for these new bits and pieces to be considered in terms of others.
More simply, the tools and formats we use all the time (spreadsheets, calendars, notepads, computers, odometers etc…) already exist but they don’t currently take into account the further levels of data we create. We don’t have a tool to see our car’s mileage at a certain date, so we’d need to walk out to the car, look at the odometer, and guess. The bit that’s missing is the connection—the link between information we have and a tool or another bit of data. In the previous example, we need a database to collect mileage, a connection between that and date data, and a calendar to view it—tools and data.
There are two sides to these software tools, though. There’s the side presented to the user, and the side that is accessed by processors and memory and software. I’ll blog more on the human-side later, but the “stuff” happens at the edge of these two coming together.
The “Semantic Web” works on a framework which enables any data to be easily connected to other data. Instead of sitting in a traditional relational database, which makes its connections based on a set of specific instructions (schemas), all the data are encoded with a bit of information identifying them to the web. In essence, each piece of data has an address, and can be pointed to much like a web site points to another. This works at various levels of granularity, so individual records can be linked very easily, allowing for applications to be written on top of these linked data. These applications can then let us analyse, manipulate, swap, and USE anything, literally, that we can link.
Alongside this linked data infrastructure (call it the Semantic Web, or Data Web or just the Web) is the proliferation of computing hardware. Processors and memory are being manufactured into just about anything we can buy. Thiese are all working to take the stuff we do and “translate” it into data. Phones, cars, fridges, credit cards, clocks, scales, watches… we’re surrounded by little processors or bits of memory recording and crunching what we do. What makes this situation currently frustrating/exciting is that they currently don’t share their information, and aren’t “aware” of the potential of other computing.
So, what am I getting at? Well, like we’re saying over on Nodalities, hook it up! We’re getting data, that’s happening. We have the framework(s) and the distributed network (the Web), and we have decades of experience automating data-comparisons (which is all Software ever does, if you boil it down).
The next step is to connect it.