Development

ZOO-Project is a WPS (Web Processing Service) open source project released in April 2010 under a MIT/X-11 style license . It provides an OGC WPS 1.0.0 compliant developer-friendly framework to create and chain WPS Web services. ZOO-Project is made of three parts:

• ZOO-Kernel: a server-side C Kernel which makes it possible to manage and chain Web services coded in several programming languages.
• ZOO-Services : a growing suite of example Web services based on various Open Source libraries such as GDAL/OGR, GRASS7 and integration on independent simulation models coded in Fortran.
• ZOO-API : A server-side JavaScript API able to call and chain the ZOO Services, which makes the implementation of complex geoprocessing workflows easy.

This presentation first aims to present ZOO-Project along with some common use cases.

The presentation will also highlight improvements that were made since the project inception and the ZOO-Kernel-1.0 release in April 2010. ZOO-Kernel-1.2.0 new features that includes GetStatus service and Session cookies management, as documentation and packaging effort (autotools improvements, Mac Installer, OpenSuse packages...).

Further, a number of sucessful experiments that were carried with new ZOO Services will be described. Indeed, ZOO 1.2 now supports GRASS7 thanks to the Python WPS-bridge, and this significantly extends the ZOO based services with a variety of advanced spatial algortithms. Some other experiments using GDAL and MapServer will be shown as an example of how complex processes can be simply chained and orchestrated using the JavaScript ZOO API.

Another interesting research using ZOO-Kernel Java support on the Cloud will also be described. The grid implementation allows to deploy ZOO-Project on the Sun Grid Engine to run complex GPS pocessing algorithms implemented in goGPS project (www.gogps-project.org).

"Geo is cool." We all know that. New companies have been created based on the fact that we all carry Internet enabled GPS devices with us at all times. As there has been more adoption of new data stores, there are also new tools for querying that data. Geocouch combined with Javascript geoJSON functions can make easy work of geodata.

The first part of the talk will report on the major developments and accomplishments in MapFish over last year, and present what's in store for the year to come.

The second part of the talk will report on good practices applied by Camptocamp when working with the MapFish framework such as:

• using MapFish web services instead of WFS when it makes more sense for the application
• setting up MapFish applications in Apache (to combine performance and robustness)
• using HTTP caching mechanisms for static and dynamic resources 
• setting up the MapFish Print system for use in the application
• setting up a common security layer for controlling access to MapFish, TileCache, and OGC Web services
• the methods and tools used to build and deploy MapFish applications
• using "application templates" to speed up setting up new applications

The open source application "MapFishSample" will also be introduced as a good illustration of the practices we use everyday when building high-performance, production-ready MapFish applications.