Why was the gvSIG Community Edition Project Started?
Originally, the development of gvSIG was funded by a Spanish administration with local and EU public funding. From 2004 to 2008, a lot of new functionality was developed by a big team of full-time, professional developers. During that period, gvSIG grew quickly. Version 1.1.2 had a lot of good extensions and was a stable software.
Since 2009, the development of the current stable version of gvSIG (release branches 1.9 to 1.11) has been decreasing dramatically, and the release of the largely redesigned version 2.0 is still pending. Some extensions made for gvSIG 1.1.2 have not even been migrated to the current 1.11 release yet, and it is uncertain whether they ever will be running with version 2.0 (in development since 2008, release was initially scheduled for September 15th 2009). There are some extensions that have never been finished or did not reach official releases status. A lot of new users have joined the project since then, but the number of full-time developers has been decreasing continuously.
Meanwhile, the next major version 2.0 is being advertised as the best option for developers. Because of the refactoring process, the structure of the code differs a lot between 1.x and 2.0. Companies that are offering gvSIG development and support services face a difficult choice, deciding which version to focus on, and potential customers are uncertain about which version they should choose to invest in.
gvSIG project contributions and official projects
More contributors are now needed for the development of gvSIG. The gvSIG project wishes to ensure the highest quality, so it was decided to establish a formal procedure for the contribution of new code and for how to add patches to existing code.
In practice, new contributors have to wait a long time to get official status (meaning that their work will be integrated into the next official version of gvSIG), because of the procedures that have to be performed beforehand and the lack of people to carry them out.
We believe that this “official” status is no longer a mark of quality but an artefact of past times, when gvSIG was developed exclusively by contractors. Given enough resources, this process could have been continued, but the overall situation has changed. We think that the development of gvSIG is being slowed down because of a lack of funding, developers and procedural overhead (which also makes the development process for external contributors more expensive).
The decisions about the development process leading to gvSIG 2.0 are still not being discussed openly in the public lists. The gvSIG team has made a commendable effort to produce English documentation, but that is not enough, as much of the critical communication (e.g. PSC List) is still handled in Spanish. The fact that non-Spanish speaking collaborators continue to be excluded from much of the project communication and the most important strategic decisions adds to the problems discussed above.
“Official” communities are now being created in Argentina, Brazil, Italy, Russia, Uruguay and Costa Rica (under construction). But nobody really knows what this means and why and when a community is official or unofficial. There has never been any kind of information or discussion in the public gvSIG mailing lists about the official constitutions of communities.
Communities (official and unofficial) are not being informed in an equal way (in a common public list)! But that is the consequence of the communication in two languages: All that is written in Spanish cannot be synchronously translated into English.
There are still people willing to continue working for the internationalization of the project, and who organize events and user meetings, do presentations, look for investments, etc. But “Official” user meetings are also being set up! If you are not in the gvSIG Association, this kind of outreach is no longer being taken into account or promoted by the “official gvSIG team”. Why should people invest in the project if they cannot understand how it works, if they do not have companies close to them involved in the project (committers, supporters, developers, etc.)?
At the beginning, the project grew without any public structure, like: SVN, bug tracker, Project Steering Committee (PSC) list, etc. (the first public PSC list appeared in 2010). In 2011 there were still no non-Spanish speaking committers involved in the highest level decision making processes.
Recently (2009), the gvSIG Association was created with the following aims (unmodified quotation): “The gvSIG Association aims gvSIG project sustainability and the development of the Free Geomatics. The Association raises a new business model around Free Software democratic values, based on cooperation and shared knowledge, where part of the generated benefit reverts on the strengthening of the gvSIG project. The sustainability of the gvSIG Project is based on maintaining the gvSIG Professional Structure and the necessary infrastructures of the gvSIG Community (www.gvsig.com).”
The gvSIG founders were discussing this new model before its creation, but there was no discussion in the Spanish and international lists about why this administrative, development and also business model should be created or supported. The official gvSIG website (www.gvsig.gva.es) was suddenly divided into www.gvsig.org and www.gvsig.com. All the project structure was migrated to the gvSIG Association.
After more than four years of working hard for the future of gvSIG in the non-Spanish speaking countries, we decided to create a new project, gvSIG Community Edition, with a decentralized organization and English as the only project language, for a more international community. Open source projects are sustained and grow via collaboration between project teams. In the case of gvSIG, this concerns projects such as NavTable, SEXTANTE, GRASS GIS and the GDAL/OGR driver libraries. For a good collaboration, it is crucial to have a non-hierarchical, transparent and open development and decision making process, where all interested parties are equally informed and involved.
Technically, gvSIG CE is a fork of the gvSIG sources under a new, more agile and open project umbrella. We will base our development work on gvSIG OADE, but we will include all improvements made for gvSIG 1.11, 1.12 and future releases. We will also focus on the best possible integration of SEXTANTE GIS, including full support for bundled GRASS GIS and SAGA ready-to-run distributions. The SEXTANTE project is closely cooperating with gvSIG CE.
We now have an international team and wish acquire new members to work with us. gvSIG CE is not intended to compete with or replace the original gvSIG. We simply wish to create a new platform for all those who can identify with the motivations listed above. We hope that both projects will benefit from a lively exchange of ideas and code. We have decided to fork the project but would like to work together as much as we can: developing in Java and under a GPL license, with an open decision process that all people involved can understand!