RM-ODP: A Critical Enabler for Sustainable Development
Achieving Sustainable Development Goals will require developing, collecting, publishing, assessing, accessing, sharing, aggregating and analyzing a huge variety -- and huge amount -- of data. This can’t be accomplished without consensus on open distributed processing architectures.
The United Nations (UN) Sustainable Development Goals Report - 2016” overview includes this: “The data requirements for the global indicators are almost as unprecedented as the Sustainable Development Goals (SDGs) themselves and constitute a tremendous challenge to all countries.”
That report is based on a proposed global indicator framework developed by the UN Inter-Agency and Expert Group on SDG Indicators. The UN Department of Economic and Social Affairs Statistics Division’s June 2016 final version of the “Framework for the Development of Environment Statistics” (FDES) represents many years of work. It is an extraordinary document that will be a key tool for collection and analysis of the environmental data required for the SDG indicators.
A massive amount of data must be gathered into the indicator databases to quantify progress. In some respects this will become easier, because the number and variety of information technologies, their SDG-related applications and the applications’ user populations are all expanding rapidly, causing an exponential increase in the amount of data produced. The real difficulty will come from parallel growth in the number and variety of different databases, spreadsheets and data streams that nations, NGOs and industry will be obliged to develop, assess, share, aggregate (and disaggregate), integrate and analyze.
An important part of the solution lies in a set of design principles for connecting the countless data systems in the SDG Indicators universe. (These principles may already be in review in the UN SDG community, but I haven’t found reference to such review in my web searches.) What’s needed, I believe, is a “Global SDG Information Systems Reference Model” based on the Reference Model for Open Distributed Processing (RM-ODP) (ISO 10746-1 to 10746-4).
Figure: Reference Model for Open Distributed Processing (RM-ODP) Viewpoints. Table from "Server Architecture Models for the National Spatial Data Infrastructures (NSDI).” Open Geospatial Consortium (2005)
People knowledgeable about enterprise information systems know about the RM-ODP. It is a widely accepted international standard for the specification of large information systems that consist of distributed and diverse components. It has played an important role in systems architecture thinking in the Open Geospatial Consortium (OGC), in most National Spatial Data Infrastructures, and in the UN Spatial Data Infrastructure.
As shown in the figure, the RM-ODP directs information system architects to look at an overall distributed information system from 5 different viewpoints. Each viewpoint needs to be developed in roughly the order shown.
The international community has already agreed on a set of Sustainable Development Goals that could provide the core of the Enterprise Viewpoint. To build out the Enterprise Viewpoint, the UN still faces the challenge of getting countries to reach greater agreement on the indicators, and the countries face the challenge of getting internal stakeholders to agree on the indicators.
The indicators and the FDES provide much of the content for the Information Viewpoint. The proposed reference model’s Information Viewpoint needs more modeling of the workflows that will provide the data that populates the SDG indicator databases. Quality assessment will be a critical workflow item, particularly because much of the data will be Volunteered Geographic Information (VGI).* So, more work remains on the first two viewpoints, but from a systems architecture perspective, these are the right viewpoints to begin with.
It’s now time to begin developing the Computational Viewpoint. The countless systems that will make up the SDG System of Systems will need to communicate, interoperate, and behave as one big distributed processing system. This can be accomplished by means of open, international, consensus derived data model standards, open encoding standards based on the open data models, and open interface standards based on the open encodings. The OGC WaterML 2.0 standard, developed in a working group organized by the joint WMO/OGC (World Meteorological Organization/Open Geospatial Consortium), provides an excellent example. Developers of the proposed Global SDG Information Systems Reference Model’s Computational Viewpoint would also need to discuss how to exploit progress in sensor webs, linked data and graph databases (as well as relational databases), Big Data analytics, and citizen science.
I was fortunate to have Bart De Lathouwer, Director, Interoperability Programs, OGC and Prof. Mike Jackson, Professor of Geospatial Science, Faculty of Engineering at the University of Nottingham, co-author with me an OGC white paper, OGC Information Technology Standards for Sustainable Development, that provides more detail and context.
I welcome readers’ comments and I welcome pointers to information that might fill gaps in my knowledge and understanding. Surely someone is already working on this.
* See “A conceptual model of the automated credibility assessment of the volunteered geographic information”, by Idris, Jackson and Ishak. Published on IOP Science. http://iopscience.iop.org/article/10.1088/1755-1315/18/1/012070/meta
Lance McKee was on the OGC startup team in 1994. He recently retired as OGC’s senior staff writer and now participates in OGC as an OGC member.