Here is another "Linked Discourse" effort via a blog post that attempts to add perspective to a developing Web based conversation. In this case, the conversation originates from Juan Sequeda's recent interview with Jana Thompson titled: Is the Semantic Web necessary (and feasible)?

Jana: What are the benefits you see to the business community in adopting semantic technology?

Me: Exposure, exploitation, of untapped treasure trove of interlinked data, information, and knowledge across disparate IT infrastructure via conceptual entry points (Entity IDs / URIs / Data Source Names) that refer to as "Context Lenses".


Jana: Do you think these benefits are great enough for businesses to adopt the changes?

Me: Yes, infrastructural heterogeneity is a fact of corporate life (growth, mergers, acquisitions etc). Any technology that addresses these challenges is extremely important and valuable. Put differently, the opportunity costs associated with IT infrastructural heterogeneity remains high!


Jana: How large do you think this impact will actually be?

Me: Huge, enterprise have been aware of their data, information, and knowledge treasure troves etc. for eons. Tapping into these via a materialization of the "information at your fingertips" vision is something they've simply been waiting to pursue without any platform lock-in, for as long as I've been in this industry.


Jana: I’ve heard, from contacts in the Bay Area, that they are skeptical of how large this impact of semantic technology will actually be on the web itself, but that the best uses of the technology are for fields such as medical information, or as you mentioned, geo-spatial data.

Me: Unfortunately, those people aren't connecting the Semantic Web and open access to heterogeneous data sources, or the intrinsic value of holistic exploration location of entity based data networks (aka Linked Data).


Jana: Are semantic technologies going to be part of the web because of people championing the cause or because it is actually a necessary step?

Me: Linked Data technology on the Web is a vital extension of the current Web. Semantic Technology without the "Web" component, or what I refer to as "Semantics Inside only" solutions, simply offer little or no value as Web enhancements based on their incongruence with the essence of the Web i.e., "Open Linkage" and no Silos! A nice looking Silo is still a Silo.


Jana: In the early days of the web, there was an explosion of new websites, due to the ease of learning HTML, from a business to a person to some crackpot talking about aliens. Even today, CSS and XHTML are not so difficult to learn that a determined person can’t learn them from W3C or other tutorials easily. If OWL becomes the norm for websites, what do you think the effects will be on the web? Do you think it is easy enough to learn that it will be readily adopted as part of the standard toolkit for web developers for businesses?

Me: Correction, learning HTML had nothing to do with the Web's success. The value proposition of the Web simply reached critical mass and you simply couldn't afford to not be part of it. The easiest route to joining the Web juggernaut was a Web Page hosted on a Web Site. The question right now is: what's the equivalent driver for the Linked Data Web bearing in mind the initial Web bootstrap. My answer is simply this: Open Data Access i.e., getting beyond the data silos that have inadvertently emerged from Web 2.0.


Jana: Following the same theme, do you think this will lead to an internet full of corporate-controlled websites, with sites only written by developers rather than individuals?

Me: Not at all, we will have an Internet owned by it's participants i.e., You and the agents that work on your behalf.


Jana: So, you are imagining technologies such as Drupal or Wordpress, that allow users to manage sites without a great deal of knowledge of the nuts and bolts of current web technologies?

Me: Not at all! I envisage simple forms that provide conduits to powerful meshes of interlinked data spaces associated with Web users.


Jana: Given all of the buzz, and my own familiarity with ontology, I am just very curious if the semantic web is truly necessary?

Me:This question is no different than saying: I hear the Web is becoming a Database, and I wonder if a Data Dictionary is necessary, or even if access to structured data is necessary. It's also akin to saying: I accept "Search" as my only mechanism for Web interaction even though in reality, I really want to be able to "Find" and "Process" relevant things at a quicker rate than I do today, relative to the amount of information, and information processing time, at my disposal.


Jana: Will it be worth it to most people to go away from the web in its current form, with keyword searches on sites like Google, to a richer and more interconnected internet with potentially better search technology?

Me: As stated above, we need to add "Find" to the portfolio of functions we seek to perform against the Web. "Finding" and "Searching" are mutually inclusive pursuits at different ends of an activity spectrum.


Jana: For our more technical readers, I have a few additional questions: If no standardization comes about for mapping relational databases to domain ontologies, how do you see that as influencing the decisions about adoption of semantic technology by businesses? After all, the success of technology often lives or dies on its ease of adoption.

Me: Standardization of RDBMS to RDF Mapping is not the critical success factor here (of course it would be nice). As stated earlier, the issue of data integration that arises from IT infrastructural heterogeneity has been with decision makers in the enterprise for ever. The problem is now seeping into the broader consumer realm via Web ubiquity. The mistakes made in the enterprise realm are now playing out in the consumer Web realm. In both realms the critical success factors are:

  1. Scalable productivity relative to exponential growth of data generated across Intranets, Extranets, and the Internet
  2. Concept based Context Lenses that transcend logical and physical data heterogeneity by putting dereferencable URIs in front of the Line of Business Application Data and/or Web Data Spaces such as Blogs, Wikis, Discussion Forums etc.).