Adopting Semantic Technology Within the Enterprise

There are some old blogs by a member of ours that discussed The Value of Tools and RDFizing data. In these blogs they illustrated how pipelining tools such as Knime, Pipeline Pilot and Sparql Motion could be used to prepare data and create RDF.

However the difficulties to prepare everything using these tools was not discussed. In fact it really was a pain. Using one tool for this and another for that meant a really inefficient, manual and time consuming process. It’s definitely not the sort of approach that an enterprise would want to deploy as part of a system architecture.

Another aspect of the approach relates to the nature of tools themselves. Firstly some of them use interpreted runtime engines; Secondly, some of them cost a lot of money; Thirdly, additional server installations are required if they were to be deployed in a system architecture. So what’s wrong with these? Well, interpreted runtime engines are never going to allow your systems to run with any decent performance particularly when you throw larger data volumes at it and business will not want to deploy systems that are not scalable, have memory limitations or fit in with current architectures.

When it comes to utilizing semantic technologies in the enterprise there aren’t too many options. We will write more on this another time but the short story is you either become Semantic Technology experts coding something yourselves using the variety of API’s available or you implement a Semantic Technology platform. Unfortunately there aren’t too many of these platforms around and they often do not play nicely with the standards you might expect, considering themselves to play the central role in any architecture. So you are being asked to rip and replace. This clearly isn’t going to encourage adoption.

We believe that semantic technologies can help solve a multitude of data related problems in the enterprise. We also believe that semantic technologies should be easy to adopt and interact nicely with traditional systems and architectures.

In4mium has looked hard at many systems and finally found a platform that we really like. It allows developers to build services, systems and architectures that can be deployed in your favourite middleware architecture. What’s more it actually creates compiled code.

We have developed a number of addons for this environment that allow you to utilize semantic technologies when, where and how you like. Whether it is a case of RDFizing data, creating REST services, creating SPARQL endpoints or designing a model driven ESB mediation layer these modules can really get you going rapidly.

Of course there is one thing we cannot provide in code and that is implementation design patterns. We have spent years working with both traditional and Semantic technologies. Design patterns across these technologies are often very different. We’ve seen this difference being one of the biggest causes of adoption failure in the enterprise.

In4mium can show you the best way to enable semantic technologies in your environment, and give you the tools to achieve it avoiding those expensive mistakes.

This entry was posted in Enterprise Data, Linked Data, RDF, Semantic Technology, Semantic Web, Technology. Bookmark the permalink.