By Stephen Putman, Senior Consultant

I begin today with an invitation to a headache...click this link:  The Linking Open Data Cloud Diagram

Ouch! That is a really complicated diagram. I believe that the  Semantic Web  suffers from the same difficulty that many worthy technologies do - the relative impossibility to describe the concept in simple terms, using concepts familiar to the vast majority of the audience. When this happens, the technology gets buried under well-meaning but hopelessly complex diagrams like this one. If you take the time to understand it, the concept is very powerful, but all the circles and lines immediately turn off most people.

Fortunately, there are simple things that you can do in your organization today that will introduce the concept of  linked data  to your staff and begin to leverage the great power that the concept holds. It will take a little bit of transition, but once the idea takes hold you can take it in several more powerful directions.

Many companies treat their applications as islands unto themselves in their basic operations, regardless of any external feeds or reporting that occurs. One result of this is that basic, seldom-changing concepts such as Country, State, and Date/Time are replicated in each system throughout the company. A basic tenet of data management states that managing data in one place is preferable to managing it in several - every time something changes, it must be maintained in however many systems use it.

One of the basic concepts of linked data is that applications will use a common repository for data like State, for example, and publish  Uniform Resource Identifiers  (URIs), or standardized location values that act much like Web-based URLs, for each value in the repository. Applications will then link to the URI for the lookup value instead of proprietary codes in use today. There are efforts to make global shared repositories for this type of data, but it is not necessary to place your trust in these data stores right away - all of this can occur within your company's firewall.

The transition to linked data does not need to be sudden or comprehensive, but can be accomplished in an incremental fashion to mitigate disruption to existing systems. Here are actions that you can begin right now to start the transition:

• If you are coding an application that uses these common lookups, store the URI in the parent table instead of the proprietary code.


• If you are using "shrink wrap" applications, construct views that reconcile the URIs and the proprietary codes, and encourage their use by end users.


• Investigate usage of common repositories in all future development and packaged software acquisition.


• Begin investigation of linking company-specific common data concepts, such as department, location, etc.

  Once the transition to a common data store is under way, your organization will have lower administration costs and more consistent data throughout the company. You will also be leading your company into the future of linked data processing that is coming soon.

photo by steve_lodefink via Flickr (Creative Commons License)

Stephen Putman has over 20 years experience supporting client/server and internet-based operations from small offices to major corporations.   He has extensive experience in a variety of front-end development tools, as well as relational database design and administration, and is extremely effective in project management and leadership roles. He is the co-author of The Data Governance eBook, available at information-management.com.