Big data best practices begin with business rules

  • Vote This Post

    1

When he first took to the stage at the World Congress on Information Technology in Montreal last week, my first thought was: Is Tom Jenkins really the best person to talk about solutions to big data problems?

Don’t get me wrong. The chairman and chief strategy officer of Open Text, Canada’s biggest software firm and a company that deserves more recognition than it gets, is no doubt far smarter than I am, and when we spoke a few years ago about his book on cloud computing, I was pretty impressed. But Open Text’s focus is on content management, and for the most part I see big data as an issue that has more to do with the kinds of applications that would sit on top of content management systems, like business intelligence, in-memory databases such as SAP’s HANA and so forth. In some respects, the rise of big data – unstructured information that increases in volume, variety and velocity – is a potential boon to companies like Open Text, because it validates the need for something to sift through and organize content. To derive real meaning from that content, though, seems to me to be beyond the realm of most CMSes.

Jenkins, however, told the WCIT audience that he has been thinking deeply about these issues for some time now, and has authored a new book, Behind the Firewall, that examines the huge amounts of information that are not findable through Google or similar public tools. He’s interested not merely in all the information snaking its way through social media but the vast repositories of data that are accumulating inside organizations. This is happening as a major shift is occurring in the way we think about information management, he said.

“We’re managing words in a search engine instead of numbers in a database,” he said, adding that narrative sciences have replaced the old models of conveying information, even of presenting information in summaries of financial data. “Many people like to see the graphs and the numeric, but a lot of people just like to read about it.”

Behind the way information is presented is the way it is treated over time by users, Jenkins said. When you publish something to the Web, for example, you basically want it to be found, usually. Behind the firewall, there are a lot more details users need to think about. Consider a typical permission tree for a document that moves through the kind of CMS software that Open Text offers, which includes just some of the considerations:

  • Indicates title of document
  • Allows download of document
  • Allows modification of document
  • Allows uploading of document
  • Allows replacement of document

“You can start to map out with great granularity the business rules you want for your documents. A great example of where this was not done? Wikileaks,” Jenkins pointed out. Of course, it doesn’t often happen this way, which is why more companies are discovering their big data problem. “This is a lot of work. It’s a controversial step in systems. What was verbal in tribal customs has to be written down.”

This, of course, is what Jenkins sees as a priority for CIOs, and he may be right. Business transformation projects based on analytics have usually fallen flat due to bad data. It’s more important than ever that IT departments start having those uncomfortable conversations about the business rules for the information that matters to them.

“We have to somehow balance between liability and the opportunity to create value. He who creates the balance is the person that wins,” he said. “Those IT professionals that strike the balance, those are the ones that do well.”

ShaneSchick ShaneSchick (22 Posts)

Shane Schick is a writer, editor and speaker who helps people create value with information technology. As Editor-at-Large with IT World Canada, Shane recently oversaw the launch of CanadianCIO.com, a new brand that empowers senior IT decision-makers through articles, videos, events, social media and a cloud-based sourcing platform. Shane was previously IT World Canada’s Vice-President, Content & Community (Editor-in-Chief), leading a digital-first strategy that included a transition from print publications to online portals and magazines. Shane regularly speaks to CIOs and IT managers at events across Canada about how they can contribute to organizational success, and comments on technology trends as a guest on CBC, BNN, CTV and other programs. A former columnist with the Globe and Mail, Shane has been recognized for journalistic excellence by the Canadian Advanced Technology Alliance and the Canadian Online Publishing Awards.


  • Bruce Stewart

    Interesting how we continue to think that narrative (which I happen to think is highly important) is somehow resistant to enumeration. Appropriate narrator signification can create the necessary data side (e.g., Cognitive Edge’s SenseMaker approach).