OS3G - Open Source, 3rd Generation

A (humble) attempt to publish news from the trenches where Free/Libre/Open-Source Software is brought to the mainstream -- and Francois Letellier's blog, too

Monday, November 27, 2006

European Task Force on ICT Uptake

Today was the final meeting of the Task Force on ICT Sector Competitiveness and ICT Uptake which I attended as an observer, after having participated in the works as co-cordinator of the working group on innovation and participant in the working group on IPR. Commissionner V. Reding, Directors-General F. Colasanti and H. Zourek participated in the meeting, along with representatives ("sherpas") of the industry and observers.

The discussions in WG2 (IPR), chaired by SAP, were animated to put it midly. The group came to the conclusion that IPR was too conversial a topic to reach a decent consensus. Some members in the group also felt that all opinions were not taken into account with equal openness. For instance facts tending to question the positive correlation between some forms of IPR and innovation in some sectors (software) were reported as beliefs or positions of a minority. On the other hand, commonplace statements such as the necessity of increased IPR protection for the sake of fostering innovation were depicted as reflecting a wide consensus.

It came to a point that ObjectWeb requested that all mentions of the organization be withdrawn from the topic paper, in order to avoid distortion of input - avoiding distortion being a legitimate request in our eyes that apparently was beyond reach.

I proposed that the following statement be inserted in the topic paper for this WG, which (to my knowledge) has not been performed in spite of support by several participants in the group: "Group members acknowledge that there's no wide consensus on the correlation between IPR protection and innovation. Studies performed by independant authoritative researchers came to different conclusions depending on the context. In some contexts, some forms of IP protection are beneficial to innovation and economic growth; conversely, other forms may instead act as an obstacle in other contexts. As stakes are high for Europe, one should be wary of common beliefs and generalizations. The concept of IPR encompasses many different realities, from brand protection to patents, and the ICT sector produces a full spectrum of products from information to hardware. IP protection has been developed during the industrial revolution and is still a reality in our world, but the way we want to shape it for the information age should be considered with care and with an open mind."

In addition to co-coordinating WG3, ObjectWeb contributed a position and proposals about openness-based innovation, which were echoed in the topic paper of WG3, but not in the final report of the ICT Task Force.

Here is this contribution for the records:

"Support emerging innovation strategies based on openness.

The term open innovation has been proposed as opposed to closed innovation to describe the process of “combining internal and external ideas as well as internal and external paths to market to advance the development of new technologies”. However benefits of openness in innovation extend beyond this. When dealing with innovation, the specificities of the ICT sector must be taken into account. One of the most salient is the production of both material/physical (e.g. consumer electronics) and information/digital (e.g. software) goods – and compounds of them. Material and information goods are extremely different and it is necessary to set up an innovation policy for the ICT sector that differentiate between these various kinds of artifacts. Emerging models of innovation based on openness leverage the ‘non rival’ nature of information goods and gain increased global impact while relying of continuous, incremental, peer production practices that are poorly captured by traditional innovation measurement techniques. Open source software is one example of collective work, often performed by individuals or companies and which delivers quality and sustained innovation without protection of creation being a central motivation.

Recommendations:
  • Academic studies of innovation based on openness should be supported through adequate funding from independent sources (e.g. through instruments such as EC-funded research projects). Appropriate metrics and measurement methodologies need to be developed. Such metrics need to be elaborated in an “out of the box” fashion, so to reflect not only direct economic impact in the ICT sector, but also the enabling effects and externalities of innovation based on openness.
  • The European IPR policy should keep a balance between the need to protect innovation and the opportunity to favour incremental innovation in an open context. As examples, the burden of proof should be on proponents of new rights and registration of prior art should be facilitated so to reduce the risk on incremental innovators; for digital goods, this may be achieved by facilitating (with regulatory and technical measures) or even automating on-line registration of prior art at no cost.
  • Technical means for remote inter-personal communication, telecommuting, collaborative online work and management should be improved. This should include public incentives which would create demand for such systems and result in a pull effect on the ICT sector itself (need for broadband networks, adequate telecommunication services, etc). Open standards play a pivotal role in the development of infrastructures software in the information society: two features are essential to the deployment of the information infrastructure needed by the information society: one is a seamless interconnection of networks and the other that the services and applications which build on them should be able to work together (interoperability).
  • When dealing with information goods, and in a very much “the fab is the lab” fashion, virtual clusters should be put in place as alternative to traditional clusters. This would bring answers to the increase of transportation costs and environmental impact and give opportunities for enhanced territory management. Such virtual clusters should be targeted to great challenges and provide the necessary environment (in terms of infrastructure and services) to facilitate the leverage of open innovation.
  • In order to leverage innovation based on openness for the benefit of the ICT sector, bridges should be built between communities with grass-root structure, academia and the business world. This may be achieved through 'meta-organizations' able to federate both individual and organizations around innovative activities based on openness. The European Institute of Technology could play an important role in this. It should include a virtual, distributed unit targeted to innovation based on openness. Activities should be structured according to the best practices of current innovative communities (e.g. open source software, Wikipedia and similar initiatives, etc). Learning, research and creation would be mixed in a single overall process, innovative production be peer-reviewed and registered as prior art in real time and made available to all in an open way, so to impact the industry and civil society at large through appropriate business models.
  • In parallel, appropriate funding mechanisms should be designed at the European level to facilitate the deployment of open innovation systems, leveraging on EU public funding and debt and equity instrument of the EIB group."
Ironically enough, my feeling as one of the youngest participants in the WG on innovation was that the group was short of ideas to innovate... in an innovative way. Apart from the above attempt, I did not see any input which would emphasize on the disruptive practices that we saw happen in the last decade(s), with the arrival of 'digital natives' both as young adults and new workers. A suggestion would be to inject younger blood in future working groups on innovation, including enthustiats, volunteers, entrepreneurs who may live far from the Commission's corridors, but much closer to where the digital society is being built, and transformed, into tommorrow's information society.

Thursday, November 23, 2006

IST2006 Networking Session on ONESSI

The IST 2006 edition of the most important European event in the field of Information Society Technologies coincided with the launch of the EU’s Seventh Framework Programme for Research and Development.

NESSI is a European technology platform aiming to provide a unified view for European research in services architectures and software infrastructures. It will define technologies, strategies and deployment policies fostering new, open, industrial solutions and societal applications that enhance the safety, security and well-being of citizens. As a founding partner, ObjectWeb contributes to the open source activities in NESSI through an endeavour called “ONESSI”.

ObjectWeb co-organized a networking session with COSS (Finnish open source competence center) about open source software and ONESSI.

Participated
to this session:
  • Morfeo, Spain
  • ObjectWeb
  • COSS, Finland
  • Linux Solution Group, Germany
  • Piedmonte Open Source Center
  • University of Szeged, Hungary
  • Politecnic University of Madrid, Spain

Tuesday, November 21, 2006

Bull Signs Partnership with JBoss

"French systems and services provider Bull has formed a broad partnership with JBoss that could see the Red Hat subsidiary's open-source middleware used more widely in Europe." (InfoWorld)

From a Bull perspective, this move makes perfect sense. But it also is an excellent opportunity for ObjectWeb to develop closer relations with JBoss and their community of users – as anticipated when Red Hat announced the acquisition of JBoss. As an independent, nonprofit organization, ObjectWeb – today, and ObjectWeb v2 soon – has an important role to play in the global open source middleware landscape that commercial entites cannot take in isolation. Open source is first and foremost about collaboration, simply because it's a rational, proven approach to build commons and to develop innovative strategies in the software arena. I would like to congratulate Bull for its long term commitment to open source and for paving the way today to new fruitfull collaboration on enterprise grade open source middleware. I also invite ObjectWeb members to welcome this announcement and explore new opportunities with an open mind.

And anyway, ObjectWeb is open. JBoss may very well join as a company. This would be an interesting message.

Wednesday, November 15, 2006

Open Source Java

Sun recently announced the distribution of Java in open source. To be more specific, it means that J2ME would be open sourced soon, J2SE in six months or so (both J2ME and J2SE Sun implementations released under GPL2 with classpath exception), and Glassfish (J2EE implementation by Sun) changed from CDDL to GPL (GPL2 w/ CP exception).

This is good news for the Java and open source communities. Distribution of J2SE and J2ME under an open source licence will undoubtedly foster the adoption of these technologies. It makes Java clearly 'compatible' with Linux - and will please all open source proponents (and users) that were a little concerned to use Java open source projects while still relying on a close source VM. Now it will be possible to have a full open source stack (e.g. Linux + J2SE + open source middleware + bespoke or OTS applications).

The choice of GPL is meaningfull, and a clear signal toward Linux - at a time when Microsoft and Novell signed a deal around virtualization technology. Java (SE and EE) can also be considered as a virtualization platform (remember 'build once run everywhere'?) from hardware independance all the way up to the Grid.

Sun has been playing with legal frameworks to ensure compatibility of open source projects to the JSRs (e.g. J2EE certification of JOnAS, JBoss, Geronimo) and now has enough experience to make sure that open source implementations of J2[SME]E remain interoperable (and compliant w/ the specs) even in case of fork. The idea is simple: you do what you want with the source code, but to call the binary 'Java' you have to have it certified. The keystone of this scheme is Sun's control over the 'Java' trademark(s).

Yet this may not be the end of the story. What will be the answer from OS communities such as Apache (with Harmony), who've been historically (and geographically) very close to Sun, and who advocate for so-called 'business friendly' open source licenses (APL*) ? What will be Sun's position wrt GPL v3?