The power of spimes
On September 30, The Association for Geographic Information's (AGIs) Big 5 event explored big data and the associated challenges, opportunities and innovations that it holds for the geospatial sector.
In the style of the Matrix, Phil invited the audience to take a leap down the rabbit hole under the influence of the metaphorical "red pill" and learn about the power of this terminology: spimes (combining space and time). As coined by Bruce Sterling in 2004, this term has more recently been linked to the Internet of Things, whereby whereby internet connected devices are producing continuous streams of data, including geospatial information and time stamps. This is the kind of data that organizations can now incorporate into their business decisions in order gain a competitive advantage.
Phil explained the difference between what the IT industry was like 25 to 30 years ago and what it is like today; the key date in time defining these two periods is February 1997.
Pre-February 1997, most of the IT problems that businesses faced were contained within the organization. By hiring IT companies (such as IBM), these issues could be solved by the IT department, which would ensure that business ran smoothly. In February 1997, the world changed as the web reached it’s tipping point and achieved mass adoption. From that point forward, business leaders augmented the intelligence held within the enterprise with additional data from the outside—typically web-based data.
In this new world, having spent time previously pulling information from data sources, we’re now moving towards push-type data technology. This is where spimes comes in, using context to derive immediate value through the content that is pushed to the consumer. Retailers have collected shopper preference data for years now with the use of geolocation (using smartphone data). Using this data, organizations can provide exquisitely personal product placement to their customers.
Bruce Sterling noted six facets of spimes; one in particular referred to mining large amounts data for things that match a given criteria, for example: an internet search engine. Having looked at the new and old worlds, Phil’s next web comparison was distribution. In a statistically normal distribution, when looking at search engines, there should be five to six top sites, however, in actuality, there are one or two as, in certain ways, the web does not conform to normal distribution. In today's fast-paced world, sites like search engines only survive if they are the best. Successful organizations aren’t restricted by the boundaries of the business; they look to their ecosystems of partners to aid their successes.
Using the data available, organizations can use their semi-structured data, unstructured data and relational databases to create powerful analytics and get ahead of the game by engaging with their customers in the most relevant ways possible. The use of the correct data, at the correct moment, in context is what is making the difference between those organizations that are fit, lean and go on to survive, and those that don’t.
Dr Phil Tetlow CEng, FIET is a chief architect for information management products, visiting professor of web science at Southampton University, executive IT architect, IBM Academy of Technology member and W3C member. You can follow him on twitter and find him on LinkedIn.