The cry for linked data began innocuously enough with the simple need to share data. It has reverberated among countless verticals, perhaps most ardently in the health care space, encompassing both the public and private sectors. The advantages of linked enterprise data can positively affect any organizationâs ROI and include:
- Greater agility
- More effective data governance implementation
- Coherent data integration
- Decreased time to action for IT
- Increased trust in data
Still, the greatest impact that linked data has on the enterprise is its penchant to interminably vanquish the silo-based culture that still persists, and which stands squarely in the way of allowing true data culture to manifest.
According to TopQuadrant Managing Director David Price, for many organizations, âThe next natural step in cases where they have data about the same thing that comes from different systems is to try to make links between those so they can have one single view about sets of data.â
And, if those links are managed correctly, they may very well lead to the proverbial single version of the truth.
From Linked Open Dataâ¦
The concept of linked enterprise data stems directly from linked open data, which has typically operated at the nexus between the public and private sectors (although it can involve either one singularly) and enabled organizations to link to and access data that are not theirs. Because of the uniform approach of semantic technologies, that data is exchangeable with virtually any data management system that utilizes smart data techniques. âAs long as we make sure that all of the data that we put in a semantic data lake adheres to standard RDF technology and we use standard ontologies and taxonomies to format the data, theyâre already integrated,â said Franz CEO Jans Aasman. âYou donât have to do anything; you can just link them together.â Thus, organizations in the private sector can readily integrate public sector linked open data into their analytics and applications in a time frame that largely bypasses typical pain points of integration and data preparation.
Such celerity could prove influential in massive data sharing endeavors like the Pentagon Papers, in which data were exchanged across international borders and numerous databases to help journalists track down instances of financial fraud. Price is leading TopQuadrantâs involvement in the Virtual Construction for Roads (V-CON) project, in which the company is contributing to an IT system that harmonizes data for road construction between a plethora of public and private sector entities in Holland and Sweden. When asked if TopQuadrantâs input on the project was based on integrating and linking data among these various parties, Price commented, âThatâs exactly where the focus is.â
â¦to Linked Enterprise Data Insight
Linked data technologies engender an identical effect when deployed within the enterprise. In cases in which different departments require the same data for different purposes, or in instances in which there are multiple repositories or applications involving the same data, linked enterprise data can provide a comprehensive data source comprised of numerous tributaries relevant for all applications. âThe difference is this stuff is enabled to also allow you to extract all the data and make it available for anybody to download itâ¦ and that includes locally,â Price commented. âYou get more flexibility and less vendor lock-in by using standards.â In what might be the most compelling use case for linked enterprise data, organizations can also link all of their data–stemming from internal and external sources–for a more profound degree of analytics based on relationship subtleties that semantic technologies instinctively perceive. Cambridge Semantics VP of Marketing John Rueter weighed in on these benefits when leveraged at scale.
âThat scale is related to an almost sort of instantaneous querying and results of an entire collection of data. It has eliminated that linear step-wise approach of multiple links or steps to get at that data. The fact that youâre marrying the combination of scale and speed youâre also, I would posit, getting better insights and more precise and accurate results based upon the sets of questions youâre asking given that youâve got the ability to access and look at all this data.â
Linked enterprise data allows all data systems to share ontologiesâsemantic modelsâthat readily adjust to include additional models and data types. The degree of flexibility they facilitate is underscored by the decreased amounts of data preparation and maintenance required to sustain what is in effect one linked system. Instead of addressing modeling requirements and system updates individually, linked enterprise data systems handle these facets of data management holistically and, in most instances, singularly. Issuing additional requirements or updating different databases in a linked data system necessitates doing so once in a centralized manner that is simultaneously reflected in the individual components of the linked data systems. âIn a semantic technology approach the data model or schema or ontology is actually its own data,â Price revealed. âThe schema is just more data and the data in some database that represents me, David Price, can actually be related to different data models at the same time in the same database.â This sort of flexibility makes for a much more agile environment in which IT teams and end users spend less time preparing data, and more reaping their benefits.
Data Governance Ramifications
Although linked enterprise data doesnât formally affect data governance as defined as the rules, roles, and responsibilities upon which sustainable use of data depends, it greatly improves its implementation. Whether ensuring regulatory compliance or reuse of data, standards-based environments furnish consistent semantics and metadata that are understood in a uniform wayâacross as many different systems as an enterprise has. One of the most pivotal points for implementing governance policy is ensuring that organizations are utilizing the same terms for the same things, and vice versa. âThe difference our technology brings is that things are much more flexible and can be changed more easily, and the relationships between things can be made much more clear,â Price remarked about the impact of linked data on facilitating governance. Furthermore, the uniform approach of linked data standards ensures that âthe items that are managed are accurate, complete, have a good definition thatâs understandable by discipline experts, and that sometimes have a more general business glossary definition and things like that,â he added.
There are multiple facets of data governance that are tied to security, such as who has the authority to view which data and how. In a linked data environment such security is imperative, particularly when sharing data across the public and private sectors. Quite possibly, security measures are reinforced even more in linked data settings than in others, since they are fortified by conventional security methods and those particular to smart data technologies. The latter involves supplementing traditional data access methods with semantic statements or triples; the former includes any array of conventional methods to protect the enterprise and its data. âThe fact that you use a technology that enables things to be public doesnât mean they have to be,â Price said. âThen you put on your own security policies. Itâs all stored in a database that can be secured at various levels of accessing the database.â
Implicit in all of the previously mentioned benefits is the fact that linked enterprise data effectively eradicates the proliferation of silos which has long complicated data management as a whole. Open data standards facilitate much more fluid data integration while decreasing temporal aspects of data preparation, shifting the emphasis on insight and action. This ability to rid the enterprise of silos is one which transcends verticals, a fact which Price readily acknowledged. âOur approach to the V-Con project is that although the organizations involved in this are National Roads Authority, our view is that the problem they are trying to solve is a general one across more than the roads industry.â In fact, it is applicable to the enterprise in general, particularly that which is attempting to sustain its data management in a long-term, streamlined manner to deliver both cost and performance boons.