Live-ish from the Esri Dev Summit

14th November 2013

I’ve spent the last three days in London at the Esri European Developer Summit. My head and notebook are full of interesting insights and technical titbits. Before this becomes an extended essay I’ll focus on what I’d tell you about over a coffee.

Geotriggers enable clever marketing but can do lots more 

I’ve heard about them, I’ve read about them, but hadn’t really seen beyond the targeted marketing use case. Amber Case’s keynote presentation was interesting but it was the tech session that she did that provided the insight. We learnt that a Geotrigger consists of a condition, an action and a tag. The condition consists of a location (point or area), whether the trigger is on entry or exit and a date/time option. Tags help with managing which users get which Geotriggers. The action can be a message, a callback url or a change of tracking profile. This last action is the one that changed my sense of what can be achieved. The three tracking profiles are primarily about battery management but they can also control the delivery rate of information. This is all coupled with callback data that comes back from the user, including which Geotrigger they activated. 

So now we could define Geotriggers for critical areas and Geotriggers for proximity areas. A crew are on a ‘quiet’ tracking profile. They enter the proximity area and receive a message. A second Geotrigger changes their profile to fine to ensure that they are immediately warned if they enter the critical area. Meanwhile the callback information shows which crews are near or in critical areas.

The developer community will get their first access to Geotriggers when the public beta opens on 19th November and it’s going to be interesting to see what uses it gets put to.

We’ll be getting an application builder for the Javascript API

The guided application builder provided by ArcGIS Viewer for Flex, and its Silverlight cousin, has proved to be very popular. How nice it would be to have the same for Javascript and get the cross platform advantage of a GIS Javascript application. So said enough users that Esri are working harder to bring this about. The approach builds on the successes and lessons learnt from the current ArcGIS Viewers but will have a different workflow. This will emphasise the intended use of the application and make it easier to tailor the look and feel. It’s early days, the beta is planned for March 2014, but getting sneak previews of what’s under development is part of what makes the dev summits so interesting.

Alongside the creation of the Viewer, 3D capabilities are being written into the Javascript API and that reminds me…

3D GIS is moving centre stage

3D GIS isn’t new to ArcGIS but the capabilities are still a little dispersed. ArcScene provides 3D capabilities on the desktop and CityEngine lets you create rich 3D landscapes. The webviewer is a great way to share 3D scenes but it sits separate to the web applications that are such a big part of GIS use now. Across a number of the sessions we got to see how 3D is becoming a more closely integrated part of the ArcGIS Platform. A new service is going to allow 3D data to be streamed and we saw how the web application is then able to draw the 3D features as they are downloaded. No more waiting for 2 minutes for a scene to load into the webviewer. More significantly it will allow people to use a web application to explore large extent 3D scenes since features will be loaded as they are needed.

We also got to see the new 3D symbology and labelling. One example was points on a hurricane track across the Atlantic. The point symbols were extruded in proportion to the wind speed at each location and labelled with the speed. The labelling engine deals with occlusion of labels as the user changes their viewing angle. We were also shown street-names in a city scape flipping from horizontal to vertical depending on the view angle.

Alongside this comes the integration of CityEngine into the new desktop application, ArcGIS for Professionals. To reinforce the significance of this Jim McKinney showed how easy it is to use a split view to create a feature in 2D and convert this to a 3D feature. For someone that works primarily with 2D, the idea of not having to learn to drive a separate application really appeals.

If that wasn’t enough to justify the section heading, Eaun Cameron mentioned the integration of 3D in his round up of what’s in the future for the Runtimes.


There’s lots more to tell but our imaginary coffee is drunk. I can’t leave without letting your know that there was a demo of disconnected editing in the runtimes and that was followed, in the Road Ahead session, by a first look at the next version of Collector… complete with disconnected editing.