The open-source release of the Ozone Widget Framework (OWF) in late 2012 ushered in a new way of creating and sharing online services. OWF provides a cloud-based environment where online content and services from different providers can all be presented in a single instance, and events that occur in one service can be used to trigger other actions, even across traditionally unconnected services and technologies. The display and event handling capabilities of the framework provide a unique canvas on which providers of online services can collaborate.
For the geospatial community, OWF promises a new paradigm in online mapping applications, as well as GIS and remote sensing analysis capabilities in a cloud implementation. Forward-thinking companies are building widgets that leverage OWF and can be combined with other geospatial visualisation and analysis services to build cross-platform applications on the fly in an interactive environment.
Originally developed as a project for the US Government, OWF was designed as a way for agencies to share information and applications in a single interface. It was initially distributed solely within the US Government and saw great successes. The 2012 National Defense Authorization Act mandated the release of the OWF to the public as freely available software for consumption by commercial and other non-government entities. Its release has been lauded as a great step by the federal government to create cohesion between the services shared by governments, commercial industry and other non-governmental organisations.
It also has great implications for international developers of geospatial software as well. The open source nature of OWF means that the software is available to anyone interested in developing widgets for use in the framework, which will help foster collaboration between businesses and countries alike.
In a basic sense, OWF is a cloud-enabled framework that can display content and services, known as widgets, from different sources in a single interface. Beyond simple display capabilities, it provides the ability to create event-handling rules that operate between these sources. Widgets can ‘shout’ events that happen within their construct and other widgets can ‘listen’ for these events, which can trigger a response from that new widget. This widget can then ‘shout’ its event for other widgets to hear, creating an event handling system that is decentralised from any of the widget’s specific design. This allows for information-sharing and custom workflow management built on previously disconnected sources of information and functionality.
The benefit here is that distributed entities can now take siloed stores of information or discrete functionality slices from the organisation and make them available to the larger organisation as a whole without having to centralise the data or servers that are holding it. It also means that the consumer can now take widgets published by non-partnered organisations and create integrated user interfaces that combine the functionality from different companies.
How does it work?
As an example, in Figure 1, we can see what appears to be a single, fluid interface involving a map display, a data catalogue provided by the Exelis Jagwire data management component, and a list of analysis functionalities provided by the ENVI Services Engine. While these appear to be part of a singly coded interface, they are, in fact, discrete widgets that are providing services and/or data from separate servers.
This interface is derived from four individually running widgets. Each of these widgets can be added as a different component to a single interface, or ‘dashboard’, which can then be shared with others as a single entity. Dashboards can also be created on the fly by individual users to create a custom interface specific to their workflow.
Here’s how it works. When a user clicks on a data source in the widget, that widget ‘shouts’ the event to the map widget, which zooms to the appropriate extent and displays the data. Clicking on an analysis task in the ESE Tasks widget shouts out to the ESE Parameters widget below it, which then displays the appropriate parameter list for that specific task. The parameters task is then listening for an event from the catalogue, where a click on a dataset sets it as the input for the task which, once it completes, sends the data source back to Jagwire and to the map display.
In Figure 2, we can see an example of a new interface, which uses the Exelis LiDAR Viewer widget instead of the Exelis Map widget. This was created by simply replacing the Map widget with the LiDAR viewer widget, and defining the way that the LiDAR viewer responds to events within the other components of the dashboard, such as the clicking of a dataset within the Jagwire widget.
In this new dashboard, clicking on a LiDAR dataset brings that point cloud up in the streaming viewer and allows the user to manipulate the dataset using the mouse or keyboard controls. This highlights one of the key benefits of OWF, which is the ability to quickly configure dashboards built from different functionality components and define the way that those components interact with each other. This greatly simplifies the creation of workflow-specific interfaces and further enables both the individual analyst as well as analysis teams within a cloud-based network.
With the recent open-source release of OWF, end-users are now able to quickly create workflow-specific interfaces from previously disconnected slices of functionality. Geospatial companies are working to build out widgets that can be shared among GIS users and data analysts to promote the integration of their technologies into this cloud-enabled, open source framework. Through the adoption of this new paradigm in web-top development by commercial industry, we will begin to see the benefits of on-the-fly, workflow-specific interface development and dissemination that made the OWF such a success before its release.
We will begin to see the benefits of on-the-fly, workflow-specific interface development and dissemination that made the OWF such a success before its release
Patrick Collins is a solution engineer at Exelis Visual Information Solutions (www.exelisvis.com)