Go to the Globe and Mail homepage

Jump to main navigationJump to main content



Semantic Web offers challenges and opportunity Add to ...

Let’s say you’re planning a wedding. You decide it will be held next June, and when it comes to floral arrangements, you want all of the flowers to be red.

Any good florist should be able to come up with a reasonable estimate for this request. After all, a florist knows which types of flowers are red, which are available in June, the various locations that red flowers are grown, and, based on past prices, how much they’re likely to cost at various times of the year. A good florist might also factor in things like weather conditions, shipping costs, or global supply and demand.

For a computer however, providing an accurate quote would be a much trickier proposition. But it’s a good example of the type of problem that the Semantic Web – a movement to add meaning to online information – could help to solve.

Codifying relationships

“The tools to create ontologies and to map your domain knowledge are getting easier and easier to use. So it’s happening, it’s just not quite here yet,” says Diane Mueller. the “cloud evangelist” at ActiveState , a Vancouver-based software company.

Her primary focus has been large-scale finance and accounting, but she also sees potential for small businesses to use semantic technologies.

Mr. Mueller says a human florist can give you a quote for all-red flowers in June, “because they’ve got years of experience figuring out that question. They know how to price out that wedding.”

In theory, a computer could use semantic techniques to do the same thing. Much of the information needed is already online.

“We know what the weather is. We know where the flowers grow. We know how much they cost year-over-year. All those datasets are becoming available.”

The problem, Ms. Mueller explains, is that there’s nothing connecting those datasets.

That’s where the Semantic Web comes in.

By creating ontologies (machine-readable explanations of the relationships between bits of data, businesses can can begin to combine information from disparate sources, making it more useful and valuable.

The importance of domain knowledge

Ms. Mueller says that, for certain domains, such as finance, accounting, and clinical trial research, detailed ontologies exist and are being used. But for many other fields, “the ontologies are not there yet. The ontologies are still in the head of the florist who’s been in the business for 20 years.”

For small businesses, this is both a challenge and an opportunity. If there isn’t an existing ontology for your field, “all that information is locked in somebody’s head.”

Creating an ontology means getting that information out of a person’s head, and into a format that a computer can understand. “Basically, the person with domain knowledge tells the software what all the relationships are,” Ms. Mueller says.

For example, a rose is a flower. A rose can be red. Rose pricing and availability depends on distributors and the time of year. And so on and so on.

According to Frédérick Giasson, a Québec City-based computer scientist and co-founder of Structured Dynamics, building an ontology is like defining your business’s world view.

And there are clear business advantages for creating these world views. Chief among them: cost savings.

Rather than hiring consultants to perform ongoing maintenance and management of a corporate database, the flexibility of a semantic system allows companies to “rely on their own team, on their own people who really know better the domain and the needs than any other external office,” he says.

Mr. Giasson believes that, for most small businesses, the people who work with data on a day-to-day basis are in a better to position to manage it than a consulting database administrator. “An ontology is really something that is better defined by the domain experts, not by programmers or IT people,” he says.

Existing tools can be clunky

While the Semantic Web offers opportunities for businesses to better use the information they already have by linking it and combining it with other datasets, there’s a challenge: Today, the tools for building semantic solutions aren’t always accessible to non-technical users.

“That’s really the big problem I see for SMBs and the Semantic Web: the tools are too costly, and the setup and the domain knowledge to create those ontologies [are] too costly to really do anything of significance at the moment,” Ms. Mueller says.

She sees that changing, though. She predicts that semantic tools will become cheaper, easier to use, and incorporated into online services. Many users may not even know they’re using “the Semantic Web.”

“I think what you’re going to see is that there’ll be a tremendous number of startups and SaaS [software as a service] offerings that have these engines embedded in them,” she says.

But, will the Semantic Web ever reach a point where it can replace a florist’s particular subject expertise?

“You’ll never replace the design sense of a great florist.”

But for tasks like generating quotes based on disparate datasets like, time, weather, and supply chains, “those kinds of things will be doable.”

Special to The Globe and Mail

Other stories can be found on the Web Strategy section of the Report on Small Business website .

Join The Globe’s Small Business LinkedIn group to network with other entrepreneurs and to discuss topical issues.

Our free weekly small-business newsletter is now available. Every Friday a team of editors selects the top picks from our blog posts, features, multimedia and columnists, and delivers them to your inbox. If you have registered for The Globe's website, you can sign up here. Click on the Small Business Briefing checkbox and hit 'save changes.' If you need to register for the site, click here.


Report Typo/Error

Follow us on Twitter: @GlobeSmallBiz

Next story




Most popular videos »

More from The Globe and Mail

Most popular