Talkstandards » Blog Archive » Shortlines: The new European Interoperability Framework

Shortlines: The new European Interoperability Framework

posted by Susannah Sheppard on Wednesday, February 23rd, 2011 at 4:14 pm

This article is co-authored by Susannah Sheppard, Consultant, and Richard Kemp, Senior and Commercial Technology Partner of Kemp Little LLP.

This comment highlights several of the issues we raised in a recently published article that discusses the recently released EIF v2, located at http://www.kemplittle.com/html/stay-posted/publications/short-lines/the-new-european-interoperability-framework.html?SESSIONFRONT=52d1177702a3926ecb525037e483b78a

The EIF v2 sets out attributes of “open specification” as well as a comment which appears to give public administrations some discretion to reference specifications that do not meet these openness criteria “if open specifications do not exist or do not meet functional interoperability needs.”( Section 5.2.1) How does this differ from the language in EIF v1 and what are the practical implications of this difference in language?

Section 5 of the EIF v2 sets out the Commission’s position on the role of interoperability agreements in creating pan-European services and recognises that on a technical level, it is inevitable that more than one specification may be able to provide the basis for the pan-European service in question. So the criteria by which authorities should decide which specification to use is defined on the basis of broad principles of transparency, fairness and non-discrimination. When there is a choice of various relevant formalised specifications, selection should be determined on the basis of:

• quality of implementation;

• market support;

• potential for reusability;

• openness

“Openness”, has been at the heart of a controversy within the software community as some proponents of FLOSS have historically equated ‘open’ with lack of license fees or royalties and in the standards context, have suggested that an “open standard” must not have any patent royalties associated with it, because they claim that patent royalties are incompatible with FLOSS licenses. EIF v2 has resolved this debate.

Paragraph 5.2.1 of the EIF v2 rejects the assertion that ‘open’ means royalty free in its contextual definition of the meaning of ‘open specification’ which accommodates both proprietary and open source software implementations:

If the openness principle is applied in full:

• all stakeholders have the same possibility of contributing to the development of the specification and public review is part of the decision-making process;

• the specification is available for everybody to study;

• intellectual property rights related to the specification are licensed on FRAND(FN1) terms or on a royalty free basis in a way that allows implementation in both proprietary and open source software (FN2).

FN1: FRAND: Fair, reasonable and non-discriminatory.

FN2: This fosters competition since providers working under various business models may compete to deliver products, technologies and services based on such specifications.

From this it is clear that a FRAND based patent license is not considered incompatible with implementation of standards in open source software.

The second footnote sets out an important policy objective lying behind this position. The Commission needs to ensure that specifications do not to unfairly exclude different types of business models. (such as; a software and IPR licensing based model; a services or subscription based model; a hardware based model and an advertising based model, the last three being business models that rely most heavily on the use of FLOSS in their product offerings).

This fits in with the general principles of public procurement law and policy , since to exclude or discriminate against large sectors of potential suppliers at the point of the tender specification would not only weaken competition, but would limit the choice of the public administration, potentially tying it, unnecessarily, to one particular model which might not, in all circumstances, be the best option. The exclusion of FRAND licensing and the insistence on royalty free standards would weaken the position of those companies that rely on a software/IP licensing model and provide an unfair advantage to those companies who have business models that would benefit from not paying IPR royalties .

Recognising FRAND opens up competition from proprietary providers as well as permitting any open source solution that does not use a FLOSS licence which prohibits the distributor from accepting a FRAND patent license, of which there are many, such as BSD, Apache and the European Public License. Even some upstream third party patent licenses with a royalty have been deemed compatible with the GNU General Public License, the one FLOSS license often pointed out as having the most tension with FRAND (see Firestar v. Red Hat, involving claims of infringement relating to Red Hat’s JBOSS product http://press.redhat.com/2008/06/11/red-hat-puts-patent-issue-to-rest/

The new wording of this section of the EIF represents a major shift from the definition of openness in EIF v1 which came down clearly against the recognition of IP rights in standards and stated that for a standard to be open it needed to comply with the following four characteristics:

1) The open standard is adopted and will be maintained by a not-for-profit organisation, and its ongoing development occurs on the basis of an open decision-making procedure available to all interested parties (consensus or majority decision etc.).

2) The open standard has been published and the standard specification document is available either freely or at a nominal charge. It must be permissible to all to copy, distribute and use it for no fee or at a nominal fee.

3) The intellectual property – i.e. patents possibly present – of (parts of) the open standard is made irrevocably available on a royalty free basis.

4) There are no constraints on the re-use of the standard.

What are the implications of the endorsement of FRAND licensing within the new guidelines on horizontal cooperation agreements, particularly in the context of standardization?

The recognition that FRAND licensing is not incompatible with ‘open’ standards in the ICT domain then falls in line with EU law in other fields such as EU competition law where the recently adopted new guidelines on horizontal agreements endorses FRAND as a basis on which IP may be incorporated in standards. As detailed IP policies have already been elaborated (such as the ETSI IPR policy) which are deemed compatible with EU law, this need to recognise IP holders’ rights is already therefore an aspect of EU competition law. In EIFv2 the Commission recognises this need for consistency with other aspects of EU law at an earlier stage in the Communication (para 3.1) where it states that the EIF must be compatible with other EU programmes and policies:

The EIS and the EIF will be maintained under the ISA Programme and kept in line with the result of other Digital Agenda actions on interoperability and standards such as the ones on the reform of the rules on implementation of ICT standards in Europe to allow the use of certain ICT fora and consortia standards, on issuing guidelines on essential intellectual property rights and licensing conditions in standard-setting, including for ex-ante disclosure, on providing guidance on the link between ICT standardisation and public procurement to help public authorities to use standards to promote efficiency and reduce lock-in.

recent DISCUSSION

Susannah Sheppard says:

In response to Rebentisch’s comment:

My point is that it is the duty of the public authority to ensure in purchasing that it gets the best value for money for the tax payer. Is is recognised both legally and from the point of view of public and economic policy that in order to do that, the purchaser needs to ensure that competition is as open as possible. That involves not excluding a particular type of business model from being submitted in response to a tender as a matter of principle. I do not believe that I am, (nor anyone else is) as a general rule, in a position to conclude that an OS, royalty free specification is always going to be better value for money for the tax payer in the long term than one which uses a technology that needs to be FRAND licensed (it may or may not be). Only the procuring authority can determine that on a case by case basis and sometimes this may only be at the point when responses to the tender are returned. But if the procuring authority is prevented from making that assessment because it is prohibited or dissuaded from opening up the tender process to FRAND solutions, it will never know whether it might have achieved a more favourable overall commercial and economic solution which would have proven better value for the tax payer.
Rebentisch says:

I am stunned by your argument: “The exclusion of FRAND licensing and the insistence on royalty free standards would weaken the position of those companies that rely on a software/IP licensing model and provide an unfair advantage to those companies who have business models that would benefit from not paying IPR royalties .” An interoperability strategy is neutral towards the objectives of vendors, except that it pursues opposing interests, it serves the general objective to strengthen the negotiation powers of public procurement agencies. You can express that by a price tag, from a procurement perspective you minimize the costs/price of services. Of course lower prices would exclude competitors who can’t offer at these prices. That is the simple rationale of a market system. A strategic level of a cost reduction approach is that you try to avoid strategic dependencies on certain vendors, thus have to enforce interoperability, on your terms.
Gerard Freriks says:

In healthcare in Europe Member States co-operate on the creation of a joint semantic resource for all to us.
In order to create semantic interoperability formal standards and this semantic resource it will speed-up developments when National projects and procurers start to make good use of the available formal open specifications.

GF
Susannah Sheppard says:

and where the public authority also concludes that certain types of interoperability require agreement on and joint use of one particular standard in circumstances where there are various different specifications that could be selected to fulfill that need, the benefit of the new approach to ‘openness’ adopted by EIFv2 is that the purchasing authority can now consider all options, including any that might involve FRAND licensing provisions, when determining which ‘standard’ or ‘specification’ to use for a particular purpose.
Gerard Freriks says:

When it comes to selection of a choice of various relevant formalised specifications it is my opinion that:
- Semantic Interoperability will not take place when there are competing different specifications.
-Semantic interoperability must be based on formal standards
- They way IT -systems technically store and retrieve data can be as proprietary as the industry wants
- But when data is exchanged we all need to agree on several unique items we all share and cherish as common resource:
a- a domain ontology so the systems can reason about the data
b- a common reference terminology aligned with the formal ontology
c- artefeacts that define what users need to document about any topic in their domain (Archetypes)
d- one common formal standard on how to document/archive data in systems and on the wire. (When these archetypes are defined as constraints on the formal documentation/archiving model interesting systems can be designed that are capable of federation)

GF

Shortlines: The new European Interoperability Framework

recent DISCUSSION

Comment on this post

About the Author

Susannah Sheppard

Talkstandards.com

An active online community where developers, researchers, policymakers and other interested parties can share ideas and collaborate on the global standards system.