The keynote speech at MarkLogic World 2012 by Tom Koulopoulos of the Delphi Group was not only interesting and insightful in it's own right but I took from it more than just the messages about 'Big Data' opportunities and the value of innovation but quite early on it became apparent to me that Tom was not just talking about Big Data per se but Linked Data too.
Although he did not mention Linked Data explicitly, Tom Koulopoulos' presentation drew attention to the importance of 'connections' - links between data and the value that has both within an organisation's data and outside to externally held data and he chose to underline this by likening it to the neural connections within the brain.
For me, it's the connections that make Big Data work and this is no more obvious than in Social Media which is where many people are looking at issues of Volume, Velocity and Variety (the three Vs of Big Data) in data processing and management. This was also born-out earlier on in the day when MarkLogic Server was seen demonstrating a range of applications that were discovering new insights at the intersection between data sets - the, if you will, latent links. By 'latent' I mean the hidden, or as yet undiscovered, links between data sets.
If there are latent links that can be discovered through searching large volumes of widely varying data in real-time, or otherwise, then it is highly likely that there are actual links, pointing within, and outside, that are worth enriching, utilising and publishing along with the raw data too.
Linked Data defines a set of guidelines/best practices/patterns, call them what you will, for publishing data that can, and should, contain links to other data and, I've found, there is no better explanation of this, and why you should do it, than in 'Linked Data: Evolving the Web into a Global Data Space' by Tom Heath (Talis) and Christian Bizer (Freie Universität Berlin).
Linked Data is finding its place in Media, Government, Scientific, Technical and Medical Publishing and by applying and adhering to these guidelines a truly huge volume of rich and varied data is growing at an unprecedented rate but which is made all the more valuable by the connections (links) that tie it all together than the sum of its volume, velocity and variety.
Some say a fourth V is the Value derived from how quickly you can process Big Data, I'd like to add to that with the notion that it is the connections you can make inside and outside of your Big Data that is where the real value lies.