How Semantic Technology Can Help Real-World IT

Loraine Lawson

I suspect it's still hard for organizations to believe in semantic technology. After all, IT is pretty busy these days, what with supporting services (external or internal), assessing the cloud, BI (business intelligence), MDM (master data management) and all the other acronyms.

 

If you follow semantics at all, you'll see that there's actually a lot of discussion about how semantic technology and movements like LinkedData will support and simplify all of those things.

 

Recently, Internet Evolution featured a short article on five ways semantic technology could affect the enterprise: Transaction processing, knowledge management, decision support, unstructured activity and the environment, which refers to the IT and business ecosystem, and not the actual environment.

 

Not surprisingly, integration plays a direct role in two of the five issues. Semantic technology promises to help resolve mismatches between systems. When you consider that companies spend between 35 and 65 percent of their budgets on integration and interoperability issues related to transactional processes alone-and that a large percentage of those problems are caused by semantic mismatches-well, you can see how semantic technology could make a huge dent in integration costs and work.

 

That's based on a survey by Semantic Arts, a consulting firm specializing in semantics. Most of the article is based on the the company's published findings, "The CIO's Guide to Semantics," which you can download and read for free.


 

To be honest, the copyright says it's from 2005, so I'm not sure why the article calls it a new survey. I looked for an updated survey and couldn't find it, but it's very possible I just missed it. Despite the age of the report, I think the Internet Evolution piece will still strike many as "new" information, given how little coverage semantic technology receives. What's also nice about this piece is it focuses less on the technicalities of semantics and more on how those can be applied to existing enterprise technology.

 

If you'd like more details, check out the full report. Although the standards discussion has probably evolved since 2005, certainly the discussion from pages 8 onward is worth reading and more relevant to those in working IT shops anyway.

 

In particular, look at two discussions:

  • Transaction processing, which covers how semantic brokers could support or even replace EAI (enterprise application integration) hubs; and
  • Decision support, which explains how semantic technology can support ETL.

 

This piece focuses on how semantics can support enterprise IT as it exists now-but for a look at how semantic technologies, particularly as it pertains to Web 3.0, could transform business, check out this fun piece on Inventor Spot, which asks if major disasters like the BP Gulf oil spill could be avoided with semantics and tagging technology. It's written in a very tabloid style, but it connects the dots between semantics and the recent proposal by W. David Stephenson, principal of Stephenson Strategies. If you'd like a more traditional piece on Stephenson's suggestion, read about it on Federal Computer Week.

 

You might want to also check out "Seven Pillars of the Open Semantic Enterprise," which focuses more on the technology layers required to implement semantics, or our past posts on the implications of semantic technology for the enterprise and data integration.

 

Of course, we're still in the early days, so it's hard for regular IT workers-and most definitely ordinary folk like me-to separate the hype from the reality, or, more to the point, it's still hard to tell what's possible for real-world companies versus what requires MIT Ph.Ds to accomplish.

 

But that's why I like to follow semantic technology, because while it's still unusual to find articles about how semantics addresses IT and business problems, it's not as rare as it once was.

 

Certainly, there are a small, but growing group of professionals and vendors working to connect the dots. They're meeting June 21-25 in San Francisco for the 2010 Semantic Technology Conference. Hopefully, as that conference unfolds, we'll see more news explaining how semantics can support and transform business IT sooner, rather than later.



More from Our Network
Add Comment      Leave a comment on this blog post
Jun 22, 2010 9:32 AM Stijn Christiaens Stijn Christiaens  says:

Dear Loraine,

Thanks for this post. Concerning your remark on real-world companies versus what requires MIT Ph.Ds, I would like to add the following remark:

Semantics is something which starts from giving a term a more elaborate description (going beyond saying that "cheeseburger" is "a burger with cheese"), adding examples, notes, involving stakeholders, adding it to a community, setting a more generic type, fixing the concept type, adding facts (e.g., Person lives at Location) as well as business rules.

See also OMG's SBVR, as well as Malcolm Chisholm's latest book at http://data-definition.com/index.cfm

In short, you are moving anywhere between an ordinary term and a full-blown ontology. Obviously, you need to go as far as you are required to go for your specific business need.

In this sense, it is clear that you can involve anyone that can read or write for some things, and require some more analytical skills (read as those found in a good business analyst) for others.

Hope this helps,

Stijn (with Collibra)

Reply
Jul 2, 2010 5:00 AM Nick Nick  says:

The advantage of using an IT support service is that it provides the customer with availability services that will help the client proactively lessen the downtime as well as adequately address service-level commitments.

Reply

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

null
null

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.