OSS Impact International Workshop, Brussels, 28 November 2006

by Ramón Casero Cañas on 14 December 2006 , last updated

Archived This page has been archived. Its content will not be updated. Further details of our archive policy.

Introduction

This workshop, organized by EUROCITIES, presented preliminary results on a study that is being carried out under direction of R.A. Ghosh at UNU-MERIT (United Nations University - Maastricht Economic and social Research and training centre on Innovation and Technology). Other main contributors to the workshop were R. Glott (UNU-MERIT) and E.-P. Schmitz (Unisys). The results of the study have been presented in the report Study on the: Economic impact of open source software on innovation and the competitiveness of the Information and Communication Technologies (ICT) sector in the EU (pdf).

The study is centred on the impact of European Union (EU) Public Authorities (PAs) releasing software fully owned by them under an open source licence. PAs considered include administrations, research institutions, universities, agencies, public companies, etc.

The workshop was brief and interesting, and worth attending. Higher and Further Education (HE and FE) institutions are PAs, and as such would probably be affected by any European Directive regarding the release of Open Source Software (OSS). Even without a European Directive, HE and FE institutions are by their own nature and tradition potentially interested in releasing OSS, and they could benefit from the experience of other PAs. Therefore OSS Watch should keep a weather eye on this and similar studies.

Overview of the presentations

The EU spends €9 billion on software annually. An estimated 12% of governments own software that could be released as OSS, and 7% already do. Although part of this software is very specialised and not of public interest, however critical for the Public Sector (PS), OSS models are still appealing in terms of interoperability, transparency, security (as the code may be vetted), privacy, avoiding vendor lock-ins, not having to pay twice for the same product, etc.

Distribution of PS software does not imply competition with industry, or appropriation of the software by third parties acting as distributors. In addition, the enlarged availability increases the value of the PS software. In fact OSS can increase competition and thus market functionality, even though improvement of local economies is not a central goal when PAs decide to develop OSS. Moreover the observed spillover effect to local economies is small.

The main problem at this point in time is not to decide whether open source is viable. Nor is there a significant technical problem regarding development, deployment and support. The main issue is legal, as there are no guidelines to help governments engage with OSS. Currently PAs need to carry their own studies in licensing and copyright issues before starting OSS projects.

It is in this context that the European Union Public Licence (EUPL) is being developed to provide a standard licence that is compatible with European Law. Its use may become part of a general legal framework that is reinforced by a European Directive. This framework needs to address European and local regulations and cultural specifics. For example, the disclaimer rejecting all responsibility that exists in most licences, like the GNU General Public License (GPL), could be at odds with consumer rights laws in the EU.

It is believed that the PS should be interested in pooling software, but in practice it is extremely difficult to get local authorities to agree on sharing costs, especially across political boundaries. As an example, the City Council of Gijón (north of Spain) seems happier buying a proprietary administrative application from SAP for €1 million rather than sharing the development cost of €6 million with other PAs, or just using a similar application that has been developed by the Local Government of Andalusia (south of Spain).

However, incentives for pooling software are growing, so many issues apart from licensing and sharing costs need to be addressed: support, project leadership, potential size of community, ownership, intellectual property, financial sustainability, insurance, guarantee of quality, procurement laws, etc.

Case studies

Five case studies were presented during the workshop, and in all of them the IT department initiated and was then central to developing the process. The OSS community model is maintained, but instead of building a voluntary developer community the PA usually contracts the work.

The case study with the largest impact is arguably the Regional Strategy on Information Society of the Local Government of Extremadura (Spain). The plan includes network access for remote areas, 1 computer per every 2 pupils in schools, infrastructure, support and software tools for small and medium size companies and for the Health System, and full migration to open standards and OSS in the administration by 2007. The PA contracted out the development of OSS, spearheaded by gnuLinEx, Extremadura’s own Debian-based GNU/Linux distribution.

Other successful case studies are APLAWS+ and MMBASE (Content Management Systems), and TULP2G (forensic framework for extracting and decoding data). In these 3 cases the requirements were set out by the PA, and the PA coordinated and developed the whole project. The impact is smaller, but interestingly a neighbourhood effect was noted. Indeed, in the first 2 cases geographical neighbours became interested with the project and contributed to it.

Finally, a case study of cold adoption was presented. Beaumont Hospital (Ireland) was forced to migrate to OSS for budgetary reasons. The migration was led by the IT department, that needed to catch up with the new systems first. In this case the impact was even smaller, as users were not involved in any meaningful way and opposed the change.

Conclusions

Brussels is definitely worth visiting, if only for its lovely city centre and tasty food -moules marinière with pommes frites and house beer followed by authentic Belgian pralines is a must for the keen tourist. As several issues tackled at the workshop also have implications at the EU level, the city was particularly appropriate for this meeting.

OSS licensing and development methods are clearly interesting for the PS, but for most PAs that is still experimental terrain, with the main hurdles being legal and political, rather than technical. At this point the issue does not seem to be whether European PAs can engage with OSS, but rather how they should do it.

Further reading

Links

Related information from OSS Watch