The subsidiarity principle applies to the Swiss public authority landscape. This leads to decentralised structures in very many areas. In contrast to centralised structures, these imply a high demand for data exchange between the information systems of the various authority levels. eCH (cf. 1) promotes the exchange of interested stakeholders (authorities, companies, universities) in order to create relevant specifications for eGovernment solutions. This report addresses the challenge of standardising data. Data along the core capabilities of public authorities Public authorities are active in very diverse and broad areas. If one wants to provide standards on their activities, it is worth doing so along their core capabilities. “eCH-0122 Architecture eGovernment Switzerland: Fundamentals” (cf. 2) provides an overview of the top level of business capabilities of public authorities in section 4.2. These capabilities are divided into:
- Management skills that support intra- and inter-agency coordination. Examples are planning, processes, analytics ..
- core technical skills that encompass the substantive agency activities. Examples are education, health, agriculture ..
- core subject-specific skills of a prerequisite nature that encompass substantive public authority activities that are essential prerequisites for others. Examples are residents, businesses, geo-information ..
- general services that support public authorities in the performance of their tasks. These include finance, human resources, business processing ..
Data standards can be created for all of these core capabilities. eCH acts as an organisation in a very needs-oriented manner. So far, two areas stand out with regard to data standards. The core capabilities of a prerequisite nature (cf. Figure 1) comprise activities associated with keeping registers. A large set of existing eCH standards falls into this area.
A high benefit can also be expected from the standardisation of general services. It focuses here on business processing (cf. Figure 2).
The need for standardisation of the other general services is low, especially since these do not differ significantly from the corresponding activities in business. Real world – semantics – interface The question arises as to what kind of data should actually be standardised. The primary expectation is that the interface between information systems should be standardised. Figure 3 shows the information model from “eCH-0171 Quality model of attribute value confirmation for the eID”. On the far right, the central elements of the interface of this standard are shown in dark blue. The column Semantics shows the meaning of the interface elements. The semantic elements typically persist in information systems (possibly distributed over many information systems as in this concrete example). Following the continuity principle, in computer science the elements of the semantics are usually named the same as those of the real world. Accordingly, the representation of the real world in models is consistently dispensed with. After all, the semantics should represent the real world. In the context of IAM standards, however, this principle is not applied. In the IAM (as shown in Figure 3), the elements of the real world are represented separately. This makes it clear that the semantic elements are information elements that must not be confused with the real world objects. During authentication, the subject must authenticate itself, not the eIdentity.
In eCH, the elements of the semantic level are often described in so-called data standards and the interfaces accordingly in interface standards. During creation, data standards usually enjoy priority over interface standards because the latter have a more restricted field of application. This can also improve consistency between different interface standards that refer to the same data. To ensure that an attribute is encoded in the same way in all interfaces, the data standard sometimes also specifies the exchange format for the attribute (e.g. in eCH-0011, section 184.108.40.206 firstName, exchange format: eCH-0044:baseNameType). The identifiers that uniquely designate a particular instance of reality are very important in interfaces. They are of outstanding importance in cross-organisational cooperation. If the instances can be designated conclusively, they are also standardised accordingly (cf. eCH-0007:cantonAbbreviationType as identifier for a canton, for example). For the other identifiers, the authorities must maintain corresponding directories (cf. e.g. eCH-0097:organisationId as identifier for a UID entity). Description languages eCH uses internationally recognised standards as far as possible. For the description of data, either the Unified Modelling Language (UML) (cf. 3) or an XML schema (cf. 4) is used in the standards. UML is used more for overview diagrams. XML schemas are used especially where the concrete specification of XML documents as they are exchanged at interfaces is the main concern. The best practice “eCH-0035: Design of XML Schemas” gives many hints on the design of XML schemas in the context of eCH (and beyond). As far as possible and available, the XML schemas associated with the standards are published in versions at http://www.ech.ch/xmlns. These will remain there until a standard is repealed. Process The Reporting Section has so far produced most of the data and interface standards for eCH. The entry into force of the Federal Act on the Harmonisation of Registers of Residents and Other Official Registers of Persons (Register Harmonisation Act of 23 June 2006) has created a very great need for standardisation. The implementation of the Act has greatly advanced the integration of the SW systems of the various federal levels and thus also the standardisation of the data to be exchanged. In order to keep up with the changes in the context, the standards are sometimes updated annually. For example, “eCH-0011: Data standard personal data” is now at version 8.1. The consistent processing of requests for change (RfC) and the adoption of the interdependent documents have in the meantime become a task in itself, on whose correct implementation a great many authorities and their software suppliers directly depend. Cross-organisational change management poses corresponding challenges for those involved. Use The creation and maintenance of data standards in the area of core capabilities with prerequisite character (cf. Figure 1) shows how complex the integration of SW solutions in federal structures can be. The wide scope of these data standards – they are ultimately used more or less intensively in all official processes – is a challenge for all those involved. At the same time, however, their broad usage possibilities lead to an enormous increase in effectiveness and efficiency in the entire public authority landscape and e-society. Last year, more than 3,000 partners at the federal, cantonal and municipal levels, and in some cases also companies, handled well over five million transactions electronically, actively using the data standards for data exchange. As in the last few years, there are signs of further growth of around one million transactions in 2014. These data are exclusively transactions processed via Sedex. Everything that is exchanged via any user interface in applications with public authority employees, citizens or companies is not counted here. All in all, the data standardisation efforts to date are only at the very beginning. The current situation of standardisation in relation to all business capabilities of public authorities shows that there is still a lot to be done.
- All eCH documents mentioned in this text can be found at ech.ch. The easiest way to access the documents is to search the Internet for their abbreviation “eCH-xxxx”.
- http://www.omg.org/spec/UML/, also as ISO standard 19505-1:2012 and 19505-2:2012.