Skip to main content

The Semantic Web

As a taxonomist and employee of a company which publishes taxonomy and ontology management software, semantics and the Semantic Web are never far from mind. Semantics is “the study of meanings” (1) and “meaning, reference, or truth” (2). The Semantic Web “is an extension of the World Wide Web through standards set by the World Wide Web Consortium (W3C)” (3) with the intent of making Internet data machine-readable by encoding semantics within data. By virtue of common data formats, Internet data becomes portable, exchangeable, interoperable, and meaningful.

The far-reaching and perhaps optimistic realization of the Semantic Web has been called into question, especially the farther we are removed from its original proposal (4). Despite being envisioned in 1999 by Tim Berners-Lee, the Semantic Web has yet to come to fruition on the scale originally positedAmong the barriers are technical limitations, but mostly the realization of the dream confronts the reality of the sheer scale of the World Wide Web and lack of consensus on meanings (5) and standards. 

The W3C states, “The Semantic Web provides a common framework that allows data to be shared and reused across application, enterprise, and community boundaries” (6). In addition to the existing challenges around the realization of a truly global and intelligent network of machine-readable and actionable interchanges, it’s that last bit “across…community boundaries” I find particularly troubling in light of a fracturing world.

The Splinternet

Despite being coined years back, as early as 2001, the term “Splinternet” as an observable phenomenon is new to meThe term, however, captures current trends and gives credence to some fundamental doubts about the realization of a truly global Semantic Web.

What is the Splinternet? As the name suggests, the concept characterizes the Internet as divided by “various factors, such as technology, commerce, politics, nationalism, religion, and divergent national interests” (7). Technology splintering includes the specialization of proprietary platforms to keep users within a particular ecosystem of one or more companies friendly to each other. For example, not only do some applications only work on Apple’s or Microsoft’s operating systems, we also see hardware specialization, especially in Apple’s specific ports and dongles. Technology divergence can be manifested actively, such as the deliberate creation of alternate technologies, or passively, such as by the rejection of standard hardware or interchange formats, such as Semantic Web standards like RDF, OWL, and SKOS.

At the state policy level, there are countries restricting Internet access or specific topics, websites, or services (8). Similar to algorithmic content filtering, the presentation of information blocked at the organizational or state level presents a distorted, incomplete version of reality; worse, a user may or may not know what has been filtered out or that anything has been filtered out at all. Although practically (and maybe even altruistically) motivated, we see this in content delivery when movies, music, books, news, or other content is geographically restricted due to copyright or privacy laws. The Internet is fragmenting at the national level, as states put borders around the electronic world just as they control the flow of goods, and even ideas, across geographical boundaries under the real or imagined guise of sovereignty and security.

In the United States, we don’t need to look farther than our own borders to see this in practice. The past four years have shown a nation divided. That division is mirrored in the online world as liberals and conservatives (or their most extreme versions) separate themselves by platform, such as the split away from Facebook to Parler by mostly conservative members. “Called Cyberbalkanization (or sometimes cyber-balkanization), [the Splinternet] refers to the division of the Internet or the world wide web into sub-groups with specific interests (digital tribes), where the sub-group’s members almost always use the Internet or the web to communicate or read material that is only of interest to the rest of the sub-group” (9).

The Un-Semantic Splinterweb? 

Just as historical periods of cross-fertilization of people and ideas have ushered in dynamic ages of new discoveries and artistic creation (though not always done with the best interests of the citizens in mind; e.g., The Roman Empire, Moorish Spain, The Renaissance, etc.), the promise of the Semantic Web has been to bridge knowledge and foster boundaryless communities. The great irony, of course, is that we see more of these bridging opportunities in the form of social media and communication platforms than ever before even as, in parallel, we see the fragmentation of communities using these tools.

The Semantic Web may potentially be thwarted by the disintegration of common protocols in favor of nationalistic, idealistic, and technological separation. Not only do we disagree on meaning, we deliberately reject common understanding and standards to enforce policies and beliefs threatened by the exchange of ideas. If there is “the insistence that technology companies obey state laws within state borders, including laws on information extraction and sharing and on permissible speech” (10), how can we hope to connect information unless that information is benign, unthreatening, factually neutral, and, most likely, dull? 

The Semantic Intranet?

Is the dream of the Semantic Web dead? To pronounce it dead would be to ignore the ways in which it is working, even if those ways are not what was originally envisioned. At the microcosmic level, ontologies and knowledge graphs are growing in popularity and use within organizations (11)Much like the supportable and maintainable personal size of social networks seem to have a limit (12), perhaps the size and focus of an organization’s Intranet compared to the expansiveness of the Internet is a more realistic scale for applied semantics. Vocabularies and content developed within a domain of knowledge is more manageable and connectable than vast cross-domain networks, allowing practical, problem-solving semantic networks to develop for purpose-driven applications. 

Of course, the Semantic Intranet is still an encapsulated universe, reflecting the ideas, culture, and priorities of a single organization, or, at best, collaborating institutions sharing knowledge. The Linked Open Data Cloud and provide grand visions of inter-disciplinary knowledge connected and shared through common protocols, but what these visions don’t show is whether anyone is using these inter-connected domains to achieve functional work within or between enterprises. In the face of the Splinternet, these interconnected domains are no longer about meaning at all when it comes to language and ideas but are instead fragmented by deliberate divides.

While a Semantic Intranet is better than no semantics at all, we still have a long road ahead of us littered with land mines, political agendas, and barriers to reach the goal of a truly Semantic Web.

  1. “Semantics.” Merriam-Webster, Merriam-Webster, Incorporated, 2021, Accessed 9 March 2021. 
  2. Semantics.” Wikipedia, Wikimedia Foundation, Inc., 2021, Accessed 9 March 2021. 
  3. Semantic Web.” Wikipedia, Wikimedia Foundation, Inc., 2021, Accessed 9 March 2021. 
  4. Whatever Happened to the Semantic Web?, Two-Bit History, 27 May 2018, Accessed 9 March 2021. 
  5. Hitzler, Pascal. A Review of the Semantic Web Field, ACM, Inc., 1 February 2021, Accessed 9 March 2021. 
  6. Semantic Web, W3C, 2013, Accessed 9 March 2021. 
  7. “Splinternet.” Wikipedia, Wikimedia Foundation, Inc., 2021, Accessed 9 March 2021. 
  8. Ovide, Shira. Copying China’s Online Blockade.” The New York Times, 1 March 2021, Accessed 9 March 2021. 
  9. “Splinternet.” Wikipedia, Wikimedia Foundation, Inc., 2021, Accessed 9 March 2021. 
  10. Malcomson, Scott. Welcome to the SplinternetTechonomy Media, Inc., 2021, 9 March 2021. 
  11. Rahman, Rashif Ray. The Semantic Web: Where is it now?, Medium, 3 October 2021, Accessed 9 March 2021. 
  12. “Dunbar’s number.” Wikipedia, Wikimedia Foundation, Inc., 2021, Accessed 9 March 2021.