correlate Rotating Header Image

semantic web

Semantic Web Kills Startups…well it could

Came across an article by Dan Zambonini, Is Web 2.0 killing the Semantic Web? The article points out a fundamental difference between Web 2.0 versus the Semantic Web, power of people versus the power of automation respectively. It also comments that the more Web 2.0 proves things easy, the more the semantic web seems complex and furthermore, seemingly impossibly unattainable. The interesting thing is this is precisely the opposite point than what I’ve been thinking lately. So what have I been thinking?

The emergence of the Semantic Web could prove a major disruption and potentially category killer to many of the web 2.0-like, innovative start-ups that have emerged building a bridge for us to the promise of the semantic web that many believe will never arrive. New players that enable data extraction, tight data integrations and shortcut mash-up platforms could potentially all go straight to the dead pool. Think players like Dapper, Fetch, Connotate, JackBe, Kapow and others.

Now some believe the Semantic Web will never happen, that publishers will never find a standard, embrace it and publish in formats where machines based integration can take place. I am beginning to question this view simply because in my view, the pace of work in the semantic arena seems to increasing each day with new pushes like Sparql, RDF and XBRL. In other cases, the semantic web acting as a disruption will be dismissed. And in others, some of the new players in the space will make the case that the existence of a true semantic web will make them stronger and more valuable players in the ecosystem. Any any of these could be correct.

But right now, my sense is if the semantic web takes off, people will begin to publish out their data and content sets in standard formats. And with that, the need for a number of intermediary players that scrape and extract content and then provide the platform layer to manipulate the information will no longer be there. Machines will be deployed to process information from a variety for a variety of data sources depending on their goals. The data will be there to use, manipulate, combine, mash and collate for new applications. This will already see emerging today, we are simply short on data availability but the time will come, it has to.

We used to think that XML would never happen, people liked their hand-coded HTML. Now we have RSS and Atom because people see the value in distributing their information for consumption. And how many players are there in the true RSS intermediary players are there now? Only a few. Could this space follow a similar path? Time will tell but I’m leaning in that direction.