Tim Berners-Lee, the inventor of the Web and Linked Data initiator, suggested a 5 star deployment scheme for Open Data. These simple classification scheme was a great success for the Linked Open Data (LOD) movement because it established common terms and makes the concepts and goals easily accessible. Here, we try to transfer this success to the domain of geospatial processes in the web. We give examples for each step of the stars and explain costs and benefits that come along with it.
Below, we provide examples for each level of the 5 star Open Geoprocess plan. The example process used throughout is 'the supermarkets on my way home', or more technically the 100m buffer around my route from work to home that is intersected with a retail dataset of supermarkets:
| ★ | make your process available on the web under an an open license (e.g. a web form where I can enter my home and work address and get an HTML list of supermarkets and the website allows the use of the process for any application)1 |
| ★★ | document your process (inputs, outputs, ...) on the web and make it usable as an API (e.g. a SOAP or RESTful service endpoint is described on an HTML page and the endpoint accepts a route in XML format and returns an ordered text list of the supermarkets’ addresses)2 |
| ★★★ | the control interface uses an open standard and the process is described in machine-readable form (e.g. a WSDL endpoint with SensorML description, accepting proprietary XML as the route input and returning a shape file with supermarket locations)3 |
| ★★★★ | process inputs and outputs use standardized, non-proprietary or open formats (e.g. an OGC WPS process accepting GML/GPX input and returning the locations as GeoJSON4 |
| ★★★★★ | all tools and algorithms to execute a geoprocess are available for re-use under an OSI-approved license (e.g. the source code of a geoprocess is published as GPLv3 and can be executed in an Open Source WPS)5 |
As a consumer ...
As a publisher ...
It's great to ... Other than writing a custom wrapper, it's hard to integrate the process into any workflows.
As a consumer, you can do all what you can do with ★ web geoprocessing and additionally:
As a publisher ...
Splendid! The process is accessible on the Web and can be used in workflows and applications by other people with their own data, however, specific implementations are required to to just that.
As a consumer, you can do all what you can do with ★★ web geoprocessing and additionally:
As a publisher ...
Excellent! The process is available via the Web in a structured way (that is, machine-readable) and now everyone can use the data easily. On the other hand, it's still data on the Web and not data in the Web.
As a consumer, you can do all what you can do with ★★★ web geoprocessing and additionally:
As a publisher ...
Wonderful! Now it's a process in the web. The (most important) parts of the process have been documented and make it reusable.
As a consumer, you can do all what you can do with ★★★★ web geoprocessing and additionally:
As a publisher ...
Brilliant! Now it's a geoprocess, in the Web that can be integrated with other geoprocesses into workflows. Both the consumer and the publisher benefit from the network effect.