Jennifer Zaino
SemanticWeb.com Contributor

Businesses’ first steps into semantic integration don’t have to revolve around adopting specific standards or complex technologies. That’s the message from Seth Earley, the president of Earley & Associates, a taxonomy, content, and knowledge strategy consulting firm, who will be speaking about building a practical semantic framework at the Linked Data Planet Conference and Expo, set for June 17 and 18 in New York.

Before even contemplating issues such as what role RDF (Resource Description Framework) and OWL (Web Ontology Language) will play in your framework, it all starts with using the same terminology around products and services in content and document management systems, ERP and product lifecycle management software, and sales force and e-commerce applications.

“It’s an endless combination of ways to express terms,” says Earley — different terms for similar features in different products or for the same features in the same product in different geographies, for example. “You get all kinds of inconsistencies across organizations — you can’t search things effectively, you can’t have consolidated reports that roll up into meaningful summaries, you confuse your customers who don’t know what or how to order.”

But while this approach circumvents some of the issues around the adoption of semantic web technology frameworks, it doesn’t mean that getting on the same page with terminologies is going to be all that simple a task.

“When people look at this from a data perspective, people just say, “Oh that’s simple, give me your reference data so I can tag something as a document type or for a region. Just give me those values. But it’s not that simple because business terminology changes all the time. Anything we do that faces the market undergoes an evolution so you can keep up with the competition, solve customer needs, and keep up with changes in technology.”

Yet the inconsistencies must be resolved — or at least surfaced so that different segments within organizations that want to retain their own vernacular for certain reasons (perhaps sensitivity to a different culture) understand that they are taking a different tack. “Recognizing when you can be inconsistent is a big piece of this,” Earley says.

Currently, Earley doesn’t see a lot of demand from the clients he works with to move directly to semantic web formats to facilitate data merging and creating a basis for data interchange.

“They’re not necessarily deploying semantic web technologies in order to realize this. These are large organizations and they are still using spreadsheets, and because they have so many legacy systems that they would have to write translations for, they don’t know if they need to go to that level of complexity to do what they do,” he says. “For now RDF and OWL are checkboxes for specifications. But are they using it? No. Do they have plans to? No.”

But they are interested in future-proofing, and want to have that option. By establishing vocabularies and relationships among entities, however, they are preparing themselves to more easily adapt to that future.

It can take years to achieve even that — and that’s a long time to keep interest high and internal buy-in alive.

“You have to show people some improvement, even if they don’t understand the whole picture,” he says. And that improvement has to be business-specific about who is benefiting and how it impacts the bottom line. For example, some companies who have done a lot of work in this area are in the textbook publishing industry, Earley says, and they can bring products to market more quickly than those who haven’t, reduce their costs, and have efficiencies of product development and product delivery that their competitors don’t. If content is tagged and organized correctly, it gives organizations the ability to repurpose content in different ways.

He mentions Houghton-Mifflin, which has a taxonomy with 15,000 terms around state standards and learning objectives and 1 million pieces of content in their content management system. “So, if an editor is trying to put together a book on biology for a state, they will be able to marry up the standards with learning objectives and then find the content that’s most relevant. It’s an incredibly efficient system for handling something that is otherwise very labor-intensive. They can get to market in weeks vs. months.”

Yet, Earley says, a problem he sees is that often when you get into conversations with experts leading knowledge management or taxonomy projects, the talk becomes esoteric and abstract — enough so that business users come out of these discussions thinking it’s easier to just build a new interface. “The people involved in these things don’t necessarily understand the linkage to business objectives and strategy,” he says, and then they wonder why their funding dries up. “You have to make that clear to your community.”

Semantic Tech & Business Conference Returns to San Francisco

Semantic Tech & Business Conference returns to San Francisco in June! Join us from June 3-7 for complete coverage of Big Data, Linked Data, Extreme Information Management, and Semantic Web. From breakthrough approaches to solving business problems to the big data implications of fast–evolving technologies, SemTechBiz provides you with an unparalleled interactive experience and delivers tangible business value. We're offering a special early rate when you register by February 17. Sign up now!