Interview with Mike Darnell of Headup - More semantic investigations

During my recent investigations into the semantic web applications that are becoming available on the internet I've started using the Headup application.

Since then I have been talking to Mike Darnell who looks after the Creative Marketing at Israel based Semanti Web and also responsible for the development of Headup.

Mike has kindly agreed to an interview on Headup and its offering to the semantic web app market.


Ed: Hi Mike, thanks for taking the time to talk to Digital Signals today.

I've been talking about a semantic year for 2009 on Digital Signals, is this something you share, the belief that this year might see some significant headway regarding semantic adoption and development on the internet.

Mike: I think that as a rule I agree. A year or two ago we heard most major web analysts and consultancies talking about 2010 as being the tipping point for the semantic web. If we define "tipping point" as the point where the semantic web is widely adopted to the extent that it becomes a concept most households are familiar with, then obviously there should be a large influx of activity leading up to this during 2009. At least as far as "top down" semantic web is concerned it seems like this is indeed the case.

Ed: Headup takes a top down approach to semantic implementation, can you briefly explain the difference between top down and bottom up?

Mike: I think that I encountered the differentiation between these two approaches to semantic web for the first time in an excellent post written by Alex Iskold for ReadWriteWeb (http://www.readwriteweb.com/archives/the_top-down_semantic_web.php). The major distinctions between both approaches, as I see them, are:

1. "Bottom Up" semantic web requires that publishers take upon themselves to annotate and properly tag all their content. The theory is that having all published content conform to the semantic web standards published by the W3C will allow and encourage communication between web services. However conforming to the W3C standards is a laborious undertaking for publishers, and so far there are no compelling case studies proving that this effort is financially worthwhile. Personally this vision seems a tad academic and utopian to me.

2. "Top Down" semantic web is a method of enabling some of the vision of the semantic web without relying on changing the data and meta data layer on the content side. So far it has been mainly the realm of enterprise. It is being moved forward by companies because they see in it financial potential.

As opposed to "Bottom Up" semantic web, the hard financial data proving this is a worthwhile undertaking is already there. It's enough to look at the trends around APIs to realize that something big is afoot. Programmableweb.com publishes a graph that displays the availability of APIs on their homepage – I suggest you look at how the graph performs for 2009 compared to its performance for 2008. It's fascinating…

Ed: I've been doing some reading around SPARQL (Sparkle), is this what you use for the application? Or would that be a trade secret?!

Mike: SPARQL is a language created by the W3C for querying RDF repositories. Instead of using SPARQL we decided to create our own proprietary SPARQL compatible querying language that is designed to act in a similar capacity with our own data method. We chose to develop our own language because we found SPARQL lacking for the "Top Down" semantic web approach.

Ed: How did you go about deciding on your data sources for the app?

Mike: We chose our "prime" sources according to two major considerations:

• Popularity – Obviously it made good sense to select for our initial launch a set of sources that have a large following because it makes Headup an appealing download for these services' subscribers. Choosing to use Facebook as a source was pretty much a no brainer here…


• Multiplicity – We preferred choosing for our "prime" sources services that inherently bring together a number of verticals. Last.fm is a great example for this case – It brings together the two verticals of music and social networking and therefore was elected as one of our "prime" sources.
  
  

Ed: Where do you see you data sources expanding to in the future?

Mike: Part of our vision for the future of Headup is to transfer the ability to add sources to our user community. This is not some distant vision for the future but a feature that our development team is already working on. Once the definition of sources becomes a community effort our expansion will move from being a matter based on intuition, speculation prediction to a path clearly defined by the wants and wishes of a steadily growing body of users.

Ed: Obviously the application is currently aimed at the social media market, can you envisage a commercial implementation of Headup and are you considering this during your development process?

Mike: We've actually already been approached a number of times by enterprises that want to integrate our semantic web engines' capabilities in their services and products.

Ed: You've mentioned developing Headup for other browsers (it's currently only available for Firefox), how's that going?

Mike: We are hard at work developing Headup's content sharing capabilities and compatibility with non-Firefox browsers is a major issue we are working on in this respect. I expect we'll be releasing this feature to our public version soon. I'll keep you updated… ; )

Ed: I personally use Tweetdeck a lot when "tweeting", have you investigated any collaborative work with any other applications?

Mike: We've created a Ubiquity mashup which you can read about here http://blog.headup.com/2008/12/headup-ubiquity-mashup, and we're always open to suggestions but as of now we have yet to find a partner for full scaled collaborative work.

Ed: Great, I'd just like to thank you again Mike for taking the time to talk to Digital Signals about Headup. I'd also like to take the opportunity to wish you all the best for the future and hope that 2009 does turn into the semantic year that we predict.

If anyone has any questions for Mike, feel free to contact him on his Twitter ID @pop_art, by our Twitter ID @DigitalSignals or via the comments below and I'll pass them along.


Related links:

Headup - Semantic browser integration - Digital Signals http://www.digital-constructions.com/blog/2009/03/headup-article-semantic-browser.html

Semantic web - The Foundations - Digital Signals http://www.digital-constructions.com/blog/2009/02/semantic-web-foundations.html

Semantic internet - Intelligent web future - Digital Signals http://www.digital-constructions.com/blog/2009/01/semantic-web-future-intelligent.html

blog comments powered by Disqus