GeoWeb Morning Keynote Final Day

by Matt Ball on July 25, 2008

Bill Gail a director at Microsoft Virtual Earth gave the morning keynote today at GeoWeb. The focus of the talk was on the transition of the GeoWeb, showing the current status and features of Virtual Earth and presenting three evolving scenarios for the evolution of the GeoWeb.

Gail detailed the content and functionality of Virtual Earth today, with its inclusion of vector maps, satellite/aerial imagery, bird’s eye views, vector overlays on bird’s eye (dotted road behind buildings), and 3D city models for hundreds of cities. The challenge and commitment is for complete, current and accurate data in 2D and 3D.

The functionality of Virtual Earth includes proximity points of interest to find items of interest within given areas, driving and walking directions, directions in 15 languages, geocoding and reverse geocoding and georss.

Ease of use is one of the key benefits of the GeoWeb, taking geospatial information that has been siloed among specialists and making it available to everyone. Microsoft offers simple integration with BI and ERP software, cross-browser support, and standard APIs for easy integration.

Microsoft has a commitment to GeoWeb applications, with ongoing innovation. And has years of experience dating back to 1995, with the first consumer MapPoint program. Acquisitions of GeoTango, Vexcel, Multimap and Caligari have enhanced their expertise.

Following are the three evolving GeoWeb Scenarios that Gail presented:

Connecting with Customers on the Web, with FedEx as an example. They built in the locations of all their stores in Virtual Earth within three weeks. Where this could evolve, is with richly discoverable locations, where you might be able to go inside the store, find parking before going there, and finding nearby points of interest to get more value from the trip.

Engaging Customers on Location, with Marriott hotels as an example. Marriott has integrated Virtual Earth in the lobbies of hotels with their GoBoard technology. The tool allows visitors to understand their local environment, with discovery of restaurants, transportation and other tourist options. Microsoft’s Surface Table might be utilized in the future, with the scenario of guests placing their camera on the surface, downloading phone images onto the table, manipulating images on the surface, georeferencing those images and annotating the images to share their experience with future travelers.

Visualizing Business Information, with the City of Miami as an example. The City teamed with Microsoft partner IS Consulting to extend GIS data to the Virtual Earth environment. With six months of development, the city posted a zoning application with property and parcel data. The tool is being used by public safety personnel and also for tourism for people to explore the area before they arrive. The future for business information is to move to cross-platforms and mobile devices to improve efficiencies of all workforces (including mobile), to share intelligence on the fly, and integrate with other models, including BIM.

Gail discussed the fact that local search is a huge market and is moving online, changing the information paradigm. Today’s reality of search is to bring information to me, but tomorrow involves taking me to the information. Perhaps you have a sense of what or where you’re searching for, but traveling to the location can greatly enhance that. If planning a trip you might go to a virtual world first and interact with local avatars to discover things that are interesting.

Gail shared a quote from John Battelle, author of “The Search” that indicates that the problem of search and discovery is at best 5% solved. With 95% of problem to be solved, there’s a lot that we can do to improve this.

Gail brought up the speech that Bill Gates gave in London on his 50th birthday, with the vision that we’ll replicate the real world in virtual reality. In order to achieve that vision, it means building out the 3D world accurately with more than 630 large urban areas, thousands more mid-range cities with the typical city having 100,000 buildings and 300,000 trees. It means a representation of a billion or more buildings.

Microsoft has tackled the problem through automation, with the right data, right algorithms, right processing technology. The first revolution of Microsoft’s 3D cities included 100 cities with10,000 buildings produced in 2 years. The next generation of cities include100,000 buildings and 300,00 trees per city that are geometrically accurate to 1M with multiple levels of detail and consistent color balancing. The automation process includes model-based object detection and placement with trees detected and analyzed for type and replaced with models. There’s also work underway to take the data down to the human level, done with photogrammetry from handheld images. Gail contends that Internet digital globes will include nearly every physical object on Earth, from global to human scales.

Gail highlighted Microsoft’s PhotoSynth technology that they’re just starting to bring online. With this tool, you can connect images created by cameras by cross-comparing images. Once you have the 3D model with photos, there’s the potential to take a picture of something on mobile phone, to pattern match that image with an image on the web, and return information about the place that puts the place in context. This could be helpful for buildings or landmarks that you’re trying to identify, or for street signs or other physical placemarks to help you navigate the environment. There’s also the potential for augmented reality, such as your car driving down street and letting you know of congestion down the road.

Gail is particularly intrigued by the merging of textual and spatial query with analytics. The thought is that when you’re visiting a location you may have a rather complex query that the concierge can answer readily, but can’t be answered by a computer. The scenario is an interest to visit a restaurant within walkable distance, with no big hills, within 10 minutes walk, with nice views, serving a particular menu at a certain price point. These detailed queries aren’t possible with a computer, but people answer these complex queries all the time. Current analysis functions such as line of sight, weather forecasts, image classification and route finding make this type of query possible, it’s just a non-trivial matter of connecting all of these. The power of Virtual Worlds is the context.

Gail stated that all is not well in the GeoWeb Paradise. We have a great vision and wonderful goals, but there’s a flawed reality to the GeoWeb. We pull together information to represent reality, but the fact is that our data is from many different timeframes with dated data altering the reality. There’s also the issue of taking 2D data to make 3D projections, with flawed reality from distorted images much like Escher drawings. There’s also the issue of orthophotos with buildings leaning in different directions.

Gail said that flawed reality can be put to good use, such as a NOAA simulation in SecondLife that allows you to see a tsunami through a simulation, to experience what you can’t or don’t want to experience in real lime. The flawed reality also lets you manipulate multiple timeframes, progressing at different rates to speed or slow down time. Combining reality and fantasy provides different versions of reality.

Would the virtual world include trash and garbage? If not, it’s not a real representation. Would it include virtual emotions in real people when avatars interact? The physiological response is something that we need to work out.

Gail sees the future of virtual worlds as reality and fantasy intertwined in interesting and elaborate ways, creating parallel connections that go two ways. Sensor and networks monitor the real world and decisions are made in virtual worlds regarding actions in the real world. Decisions in virtual world affect the real world, but there is both reality flux and fantasy flux. The question becomes where is there equilibrium between fantasy and reality, and what happens to both? The future of the GeoWeb involves the integration of fantasy and reality as we move forward.

Read more related Spatial Sustain posts:

Leave a Comment

{ 1 trackback }

Previous post:

Next post: