Geospatial Semantic Web Blog

Salvatore Salamone at InfoWorld.nl wrote a nice article that talks about the current mashup phenomenon. Salamone points out that mashup applications are not limited to just aggregating geographical data and maps. Many life science databases also can be used to combined with other information.

An example cited was of the mashup iSpecies.org. Upon entering a species into what looks like a regular query search line, the mashup returns a page with NCBI genomics information, Yahoo images of the species, and articles culled from Google Scholar.

In addition, Semantic Web technology such as RDF can help to expand the amount of data available for mashup applications. It makes easier for machines to discover, search, save and share information.

Source: Hip hop offers data integration lessons

Prof. Tim Finin posted a video of our first Semantic Web application: ITTalks. It’s a talk announcement system that publishes web contents in both human and machine readable format.

This system was developed under the DARPA DAML program. Part of the existing Web Ontology Language OWL is based on an ontology language (i.e., DAML+OIL) developed under this program. In 2000, semantic web research was mostly focused on how to publish machine readable web contents and annotate web information using shared ontological vocabularies.

If you watch the video and compare the ITTalks system with some recent Semantic Web applications, you see obvious difference in their research focus and feature capabilities.

From my personal perspective, since I have worked on the ITTalks project and still in the Semantic Web field, I think Semantic Web research has evolved from how to define ontologies to how to reasoning with ontologies, from how to publish machine readable contents to how to integrated machine readable contents and their semantics, and from how to build proof-of-concept systems to how to develop commercial quality systems.

I’m looking forward to another five years of Semantic Web development.

Jeff Thurston at Vector One writes,

“I think interoperability is not THE issue. The issue in the GIS and geospatial arena is structural barriers internal to enterprises. It is all about breaking the barriers between people.

We need to get back to the basics, discussing what it is we want to do with the many high quality tools we have created across this industry and how they can really be used - and - changing the structural processes in organizations to make it happen.”

I think interoperability is a problem, at least from the knowledge integration point of view. Organizations have invested a lot of time and money to collect information. Different information collected over the years often is not stored in the same representation format.

Though his suggestion makes great sense, I doubt that it’s the most cost-effective solution. It’s not always economical to reconsolidate datastores just because we want different schemas to align and use shared vocabularies. NGA people told me once, “reconsolidation is expensive”.

I believe semantic interoperability is a better solution.

Slashgeo points us to “The Emergence of GeoSensor Networks“, an article written by Anthony Stefanidis at Directions Magazine. In this article, Stefanidis briefly overviews the history of sensor networks and its underlying technology. He also discusses the emergence of some special sensor networks known as the GeoSensor Networks — sensor networks that are designed to collect, monitor, analyze and aggregate geospatial data.

After reading this article, I thought about some research issues that I had stumbled across few years back when doing my PhD. Those issues I think are worthy of a short discussion here.
Read the rest of this entry »

Most people have stopped arguing about why the Web needs semantics. However, a new debate is on the rise — what’s the right approach to bring semantics into the Web?

At present, there are two schools of thinking: (1) the Semantic Web (the upper-case semantic web) and (2) the semantic web (the lower-case semantic web).

Read the rest of this entry »

No wonder the US government is interested in data interoperability. Reported in a Direction Magazine article, “the cost of geospatial data non-interoperability in the US was reported to be about $14 billion/year.”

In other words, every year $14 billion is wasted because we failed to avoid “redundant data collection, costly data conversion and inability to share data developed with different data models”.

The Fourth Semantic Interoperability for E-Government Conference was held today at the MITRE Corporation (McLean, VA). The purpose of this conference is to bring together semantic interoperability practitioners to discuss and showcase technologies that can help government agencies to improve information sharing, knowledge management and interoperability. The conference expects about 250 attendees from different government agencies, universities and industries.

Colleagues and I represented Image Matters LLC to showcase the company’s new product called KnowledgeSmarts. We had an exhibition booth set up during the networking lunch session. It was fun meeting people from different backgrounds who share a common interest in semantic technology.

Read the rest of this entry »

Geospatial Semantic Web

The most value asset in a GIS system is data. Without data, a GIS system is like a computer system with the best peripherals but only with an empty hard drive — it’s useless. If data is so important, it is necessary for us to understand the role of data in the future GIS systems — i.e., the Geospatial Semantic Web.

Read the rest of this entry »