On the Difference Between "Semantic" and "Semantic Web"

This is a brief post with one purpose: to clarify the meaning of the term "semantic." It has suddenly become chic to label every new app as somehow "semantic" but what does this mean really? Are all "semantic" apps part of the "Semantic Web?" What is the criteria for something to be "semantic" versus "Semantic Web" anyway?

It's pretty simple actually. Any app that can understand language to some degree could be labeled as "semantic." So even Google is somewhat of a semantic application by that criterion. Of course some applications are a lot more semantic than others. Powerset is more semantic than Google, for example, because it understands natural language, not just keywords.

But for an application to be considered part of the "Semantic Web" it has to support a set of open standards defined by the W3C, including at the very least RDF, and potentially also OWL and SPARQL. These are the technologies that collectively comprise the Semantic Web. Supporting these technologies means making at least some RDF data visible to outside applications.

I'm not sure if Powerset is doing this yet, nor whether Freebase is doing it yet, but they should (and I'm guessing they will). Twine, my company's application, is using RDF and OWL internally within our app and we are also exposing this via our site (although we are still in private beta so only beta participants can see that data today). Other companies such as Digg are already making their RDF data visible to the public. Any application with at least publishes RDF data can be considered to be both semantic and part of the Semantic Web.

Comments

[Richard Hopkins]

Whilst I have no particular axe to grind on Semantic vs. semantic, we're beginning to build systems that are constructed from the ground up using semantics (with a big 'S' I guess). Now RDF, RDFS, OWL and the like are rather useful for doing that kind of thing, so I applaud Nova's enthusiasm - but until I can offer my clients Semantic Web capabilities for virtually nothing, they will not do much for the cause... If we begin to build systems using these standards though, perhaps they could get such capabilities for almost free?

[David Scott Lewis]

How's this for an argument, Nova: What's important is the "semantic web," NOT necessarily the "Semantic Web."

In other words, what is most important is that something works, something provides a solution, something meets a consumer/customer need. The "Semantic Web" will eventually help with interoperability. This is important, to be sure.

But with hosted SaaS solutions, as long as there is a SOA framework for integrating SaaS solutions with siloed packaged software, then everything is (mostly) on the right track.

Sure, the Semantic Web as standards does help, but I wonder (and question) whether such standards are as necessary with SaaS solutions.

The other thing that concerns me is that standards move slower than a tree grows. I've had the unfortunate experience (actually, it was kind of fun, but not terribly productive) to sit on some IEEE standards committees. The technology was at least a few years ahead of the corresponding standards. Interoperability was the key issue. But if someone wants a solution that doesn't necessarily have to be integrated with their existing solutions, or if they can apply "SOA 101" and create a workable solution, isn't the "semantic web" (as a solution) being impeded by the "Semantic Web" (as standards)? Just a thought.

[Jilles van Gurp]

I think this is a rather narrow/elitist view to tie the technology to a few standards that in my view continue to have problematic adoption in the real world where there is an abundance of mineable data.

Semantics of (S)semantic is a very unproductive discussion in my view. In my view there is datastructures, ways to convert between them, ways to mine information from unstructured data and most importantly, technology to reason with data. If the data is not in the right format, parse & transform it. Transforming graphs from one format to another, should not be an issue.

The semantic web not being a part of the Semantic web is therefore a problem of the Semantic web tools and not the other way around. The solution is to do what Google does, just deal with it and parse, convert and transform the data as you encounter it into whatever you need instead of trying to coerce the rest of the world to store and exchange information in certain ways.

It's a strategy that works and I think in terms of algorithms to do the really interesting work of extracting and manipulating meaning they are not that far removed from what Semantic Web companies are doing.

W3C based standards have a definite role in helping open up what Google calls the deep web. But so do other technologies such as Atom, RSS, Microformats, KML and the many other dialects out there to transmit information.

[David Scott Lewis]

Then perhaps what is needed is a "semantic web," furthered by "Semantic Web" standards.

But the key point is having a "semantic web," not following what a standards body decides. In other words, making it work, rather than agreement by a committee.

I've been on my share of IEEE standards committees. There was always less concern about making something work versus making something compatible. Although compatibility is certainly desirable, it's more important to make something work.

For example, a semweb end user couldn't care less about the standards: They just want something that works. (True for just about all core technologies underlying a standard.) And with a SaaS-based solution, following standards may be less important as long as their is a SOA framework for integration.

Bottom line: Whether Freebase or Powerset uses "Semantic Web" standards is far less important than whether they're helping create a "semantic web." (This is a relative statement.) Don't get too hung up on standards, especially with a SaaS-based solution. APIs and connectors can certainly do the trick; you don't need the same level of interoperability as one might with siloed apps.