The Semantics of the Semantic Web: Don’t Confuse the Concept with the Movement
By Jeff Sayre
This past week, another rash of the “Semantic Web is dead” cries have made their rounds at conferences (at least one) and across the Twittersphere. This is sensationalism at its best—and used quite effectively.
The truth is that the promise and accrued benefits of the Semantic Web are far from dead. The Semantic Web is growing as we speak. Facebook’s Open Graph ontology clearly demonstrates that big social networks are beginning to understand the power and benefits of opening up some of their data to the Web of Data.
What this bold proclamation–which has once again sparked a vigorous debate–clearly shows, however, is that the Semantic Web has a big issue with its own semantics.
It Hurts to Step on a Fork in the Road
When some people speak of the Semantic Web they understand that to be different than Linked Data. Whereas, from a technologist’s stand point, there are salient differences between these two terms, for the vast majority of netizens they could care less about the engine behind the machine.
The reality is that these two terms are different movements pushing for the same outcome. I won’t go into the politics, policy differences, and technological disagreements that led to the fracturing of the Semantic Web umbrella into multiple camps, into what are best called differing movements. Suffice it to say that whether technologists like the term Semantic Web or not, it has made its way into the mainstream lexicon.
One reason that bold statements like the “Semantic Web is dead” periodically resurface is that the higher-level concept and promise of the Semantic Web have become lost in the underlying movements. Members of the Semantic Web umbrella have hampered their success by placing too many forks in the road. They have created a road that to some people seems undesirable to travel.
To the basic Web user, the implementation details of a semantified Web do not and should not matter. What does matter are the accrued benefits, whether these benefits are perceived or not by the users. Discussions about which technologies are best at bringing a fully-actualized Web of Data to fruition should be done out of the limelight.
Rebranding the Semantic Web; Getting the Semantics Right
When I use the phrase Semantic Web I use it in the broadest sense of the concept. I do not use it to plant my flag in one camp, to pledge my allegiance to one set of technologies, to cozy up to one particular movement. To me, the terms Semantic Web, Linked Data, and Web of Linked Data are the same. They are phrases that represent a powerful concept, a noble and beneficial goal. They all connote the Web of Data.
In a previous article, I wrote about the need to repackage the promise of the Social Semantic Web. My point in essence was that in order for others to better understand and appreciate the benefits of the semantification of their data, a major rebranding of the term “Semantic Web” was required; a major marketing and educational effort was necessary.
As I state in this reply to a comment on my Apple’s Ping Versus the Social Web article:
The marketing of Semantic Web concepts, technologies, and benefits was somewhat bungled from the start. Unless you were a technologist, the packaging of the SemWeb brand was about as enticing as a box of overly-ripe bananas. No one would want to open that box.
This rehashed debate demonstrates that there is a need to rebrand the Semantic Web, to recapture its true and overarching meaning.
If the Web of Data is to become a healthy reality, we must move beyond the nitpicking semantics that pigeonholes the terms Semantic Web and Linked Data into specific boxes of technologies, into two camps at odds with implementation details.
Whereas some use the term Web 3.0 as a better moniker for the Semantic Web brand, Web 3.0 is more than just a synonym for Linked Data, for the Web of Data. Perhaps the terms Semantic Web and Linked Data are too stale or damaged to repurpose as a reinvigorated brand. Perhaps we should simply use the most apt and to-the-point descriptor we have—Web of Data.
As it is predicted that more data will travel across the mobile-based Internet, and not via Web instances, perhaps Web of Data is the better term. But the semantics of the word Web will need to be clarified. Interconnected data located on the Internet that are machine discoverable and understandable is analogous to the Web of Life in ecology. The “web” means the connections, the interdependencies. Therefore, the Web of Data is not about the Web-based, browser-accessible portion of the Internet. Instead, it is about the global connectedness of data.
Whichever term is chosen for the rebranding effort, the key point is to provide an effective, consistent, and clear message. For instance, the physics of aerodynamic lift and the mechanics of a plane do not matter to (most) passengers. What matters is that they get to their destination safely and hopefully on time. Airlines cannot run a successful brand campaign by extolling the virtues of lift. Instead, they skip the details and focus on the benefits. Likewise, focusing on the underlying technologies used to link data (and the differences in approach) does not a successful marketing campaign make. Instead the message must focus on accrued benefits to individuals, companies, and society.
Is the Semantic Web dead? That all depends on whether you’re referring to the concept or the movement. Whereas it may very well be that it’s time to put the phrase Semantic Web to rest, the ideals, goals, and vision of the original meaning are alive and well.