Esri Federal Users Conference 2013: GIS Is Evolving…Again

By Art Kalinski, GISP

Several weeks ago I attended the Esri Federal Users’ Conference in D.C. Not surprisingly, the conference was opened by Jack Dangermond, founder and president of Esri. As with most Esri conferences, there is too much to cover in this short column, so I’m providing my highlights.

Fortunately, the plenary session videos are available online. The videos are a good way to view and sample the technology and get a feel for the conference direction.

Esri Making a Difference Award

Jack opened the conference by presenting the Esri “Making a Difference” award. This year it was given to the Army’s Arlington National Cemetery (ANC) and “make a difference” they did! Shown here accepting the award from Jack is Kathryn Condon, the Executive Director and Major Nicholas Miller, the Chief Information Officer (left).

Several years ago you may remember the scandalous news reports of very poor management and almost total lack of accountability at the ANC. Countless burial plots were misrecorded, finding a burial location for visiting family members was tedious, and some tombstones were even found dumped in a nearby stream. Tracking and management was accomplished via hand-typed index cards and colored pencil annotation of paper maps. We GISPs knew instantly that GIS was the answer and fortunately so did the Army. Their solution was thorough and elegant.

The Army, working with GIS, Inc., spent several months collecting more than 250,000 point locations of grave markers across the 624-acre cemetery. They used Trimble handheld GPS units to record each tombstone to an accuracy of three inches. Additionally, a digital photo of each tombstone was taken and recorded in the database. The result is an online system called ANC Explorer, which displays GIS maps, aerial imagery and digital photos of ANC that visitors can use to navigate the cemetery. Additionally, family members who can’t visit the cemetery in person can do a virtual search and tour.

Jack’s Vision

Jack outlined his long-standing vision of where GIS is and where it is heading. Much of his vision in the past was just that, a “vision.” However, reality has caught up. He spoke of the turning point brought about by cloud computing. He used the term “GIS as a platform” where software such as ArcGIS Online is mashed with other systems to create a pervasive geographic system that enables countless applications. Web maps point back to authoritative data sources rather than carrying the baggage of large local databases. Lightweight, agile applications are extendable, interoperable and permit easy collaboration. As part of this, the emphasis is on web maps as the new medium. He stated that web maps are simple, powerful, and cost effective for collaboration and integration. ArcGIS is the platform enabling this with three components: applications, content, and infrastructure.

To make all this happen, Esri is spending considerable time and effort not only on the technology, but to assemble extensive authoritative content. This is the big difference between Esri and Google. Google has changed how we all view computer mapping, but as many of us have learned over the years, Google is not necessarily an authoritative source. By its own admission, Google’s purpose is to attract users and drive them to advertising. They have no desire to become the world’s GIS expert. Esri on the other hand is working toward that goal. One example supporting this effort is Esri’s Landscape Analyst, which is being released this month. Similar to Community Analyst, designers can develop maps based on a variety of authoritative public data sets related to land use, land cover, pipelines, transmission lines, imagery, and many other features. The interesting aspect of Landscape Analyst is that more than 30 selected data sets can be used, downloaded, and incorporated into local applications.

Another announcement that caused a buzz was the free availability of DigitalGlobe World Imagery. The new update includes 30-cm imagery for the continental United States down to 1:1,000 scale, and 60-cm imagery for large parts of Western Europe down to 1:2,000 scale. Throughout the year, Esri and DigitalGlobe will grow the coverage with more than 100 million square kilometers of updated high-quality imagery, making it one of the most detailed free basemap services available.

CityEngine

For those of you interested in 3D GIS, CAD, BIM models, Voxels and visualization, CityEngine seems to blur the lines between each and merits a serious look. CityEngine was demonstrated during the plenary session, and is clearly a very fast and easy-to-use 3D modeling environment. It combines rapid 3D model creation from 2d GIS data, similar to Google SketchUp with serious database connectivity in a procedural-based environment.

Although I personally haven’t worked with CityEngine, the demos are impressive. See a video demo. CityEngine integrates standard file formats, including Collada, RenderMan, Google Earth, SketchUp, Wavefront, 3DS, Autodesk FBX, and WebGL. The CityEngine models can be viewed in ArcScene and WebScene. An interesting capability that could be very valuable for public meetings is a slider bar that can show inside/outside or before/after images.

One word of caution that I learned working with first responders is the use of textures or cloned images to create 3D models. Previous work I did with Pictometry and PLW Modelworks highlighted the difference between a representation of a 3D scene and use of actual geo-registered imagery. PLW builds 3D models that are “photorealistic” and “photo-accurate.” The models are built from imagery and each building is draped on each side with the actual measurable image of the building. As you can imagine, it’s important that a window shown 5 feet above ground is actually there and not just a graphic. If occlusions block some portion of the image, PLW makes no attempt to fake the image. It instead shows a no-data area that looks like a black shadow.

Note this example. The “shadow” is really a no-data location of the image.

Social Media

A pervasive topic that was part of numerous sessions and Expo exhibitors was social media and big data. The integration of this real-time data with GIS data and analytics is proving to enhance both, and its use is growing. There was a powerful example shown in the plenary session using Topsy technology. Topsy uses complex algorithms to do expert searches on thousands of tweets to discover information based on keywords or terms. The search ranks results based on topical focus and geography. Since only 1 percent of tweets are explicitly geotagged, Topsy developed machine learning methods to infer an author’s location, using features such as regular references to landmarks or events. By doing so, it claims to have high-confidence geographic information for more than 90 percent of tweets. The company demonstrated the benefit of plotting real-time tweets about power outages as a second, validating source of data during Hurricane Sandy.

Several exhibitors showed social media exploitation tools that were equally impressive. TerraGo demonstrated GeoXRAY, a system that combines GIS data with social media discovery and exploitation tools, brought together into a GeoPDF that can be used off-line in local collaborations by non-GIS users.

Other exhibitors were using social media for intelligence applications, financial analysis, and business and marketing applications, with a dizzying array of claims. My personal observation is that the social media environment is like the “Wild west.” Similar to the early days of GIS, there are competing products, repackaged data, and capabilities with overlapping systems. It will take a while to sort it out, but this is too important a topic to take lightly. Look for more information in future columns as I get educated.