Direction Magazine’s Diana Sinton recently spoke with Jeremy Morley of the Ordnance Survey, Great Britain’s official mapping agency. Since its establishment over 220 years ago, the OS has contributed its cartographic expertise to the military, political, civil and social development of Great Britain. In this interview Morley touches on the role of research and education by the OS and imagines what roles digital navigational data will play in the future.
Q: Your title at OS is Chief Geospatial Scientist. Is that a position that the Ordnance Survey has long offered? What are the job responsibilities and expectations? Could you describe a typical day, if there ever is one?
A: The title of the post at least is new. Ordnance Survey has been engaged in research internally and with the academic community for decades, and so has had a research manager to run that research. The new post aims to increase the visibility of the role.
For a number of reasons research has been restructured inside OS: We run innovation, R&D and research projects inside different divisions of the business, for example, examining innovative applications of the latest technology to deliver our products and services. My job is to engage in longer-term research, generally on a 3- to 7-year horizon, though we do carry out shorter-term projects too. This includes building up our capacity internally; working with and commissioning research with external partners, especially universities; and discussing and promoting our research interests with funders and associations around Britain. We engage in research to solve particular identified problems in collecting, managing, deriving or delivering our products and services. We also invest in future-oriented research, to understand future requirements and interfaces for our information – for example in smart cities, the Internet of Things and autonomous vehicles. An important expectation of this is that while we aim for high academic quality to the work, this research has to have effect inside the business, so knowledge transfer or spin-in of the results of the research is essential. This need not necessarily produce a new product per se but may result in new capabilities to create new products and services.
Some days, therefore, involve interacting with colleagues around the business, from the Commercial division who are in contact with customers, partners and the market, through Operations who run the factory to collect, manage and derive data, through Products and Innovation who define and create not only new products but also new platforms for delivery. On other days, I will be visiting university partners to discuss collaboration on existing and new products, or presenting our research or interests at workshops and conferences.
In addition to research, my team is interested in education with a span from early age education (“primary schools” in the UK) all the way through to supporting post-graduate (master's and doctoral) education. We have collective data agreements to license and deliver our data to the school, college and university sectors in collaboration with our partners at the University of Edinburgh’s EDINA Digimap service. We also aim to support the development of school curricula in Great Britain to recognize the importance of geographical thinking and GIS, and to support teachers in delivering effective education in these areas.
Q: For many people who may not be particularly current with the Ordnance Survey, the traditional impression may remain that it’s a very closed environment, with extremely strict guidelines about access and distribution of their authoritative national data for the UK. If the culture or practices have changed at the OS in the last few years, in what ways and how?
A: Access to and licensing of OS data has changed greatly in the recent years. Most of our medium and small scale data products are now available as open data, with a standard UK Open Government License. We have for many years provided easy and low cost access to products for research and education too, as I discussed above. And we’re busy working on new licenses and means to access our data to make experimenting with products and adopting them easier, including new APIs due soon. On top of this we’re working on SDKs to work with our data. With the release of these new access mechanisms we will be introducing a freemium license model, meaning that developers and home users will be able to access even our large-scale data in limited quantities for free. As is usual in the industry we will begin charging when the volumes of data being accessed exceed certain thresholds. We hope that these developments will further enable customers’ and citizens’ access to our data.
Q: Two big trends in the world of geospatial data are the growing popularity of, and reliance on, open-source software and the incorporation of crowd-sourced solutions to data development and curation. How have either of these been incorporated into the OS?
A: We are interested in both these areas. However the biggest impact for us has been from open source. Most of our core systems still rely on the power of vendor software solutions but equally, we do use open source elements too. A mark of our interest in these technologies was our sponsorship of FOSS4G in the UK in 2013 at the top, Platinum level.
Crowd-sourcing is interesting but is not something we’ve adopted as a core part of our data collection system so far. Our products and reputation are built on high quality, consistent, national coverage and we’re still working on how crowd-sourcing best fits in that environment and best complements our use of surveying and photogrammetric data collection. An element of this to explore is the expert crowd, where we train or work with a limited pool of experts to provide information.
Q: How have you been able to integrate your own background as an educator of geospatial sciences and technology into your current position? How about your own research ideas?
A: I have previously worked in MSc education in Geographic Information Science at both University College London, where I ran their program at the end of the 90s, and more recently at the University of Nottingham. I helped develop and then ran a specialized undergraduate course at UCL in Geospatial and Environmental Information Management, which led into integrating geospatial engineering content into UCL’s Civil Engineering degrees. As a member of faculty at both universities I’ve directly supervised over a dozen PhD students and been involved in educating first-year digital economy PhD students in Nottingham’s Horizon Centre for Doctoral Training. This range of experience is invaluable in understanding the constraints and operation of GIS and geospatial education in the UK, and what faculty staff will find useful to support their teaching at different levels.
My research experience firstly means that I understand the measures of success that motivate faculty staff in engaging in research with us and the funding landscape within which we can either bid together with universities or provide support to their bids. My move to Ordnance Survey was motivated by the compatibility in my research interests, in the technical infrastructure of interoperability and internet services, in the characteristics of crowd-sourcing and the human factors in geospatial systems, and the role of GIS in the new world of smart cities.
Q: What upcoming activities or projects at the OS are you excited about? Which ones give you trepidation?
A: A really exciting area that we’re exploring is that of connected and autonomous vehicles. This is a rapidly developing area of technology which promises to radically affect patterns of transportation and the market of cars and vehicles in general. What will be the role of geospatial information in these systems? Will autonomy be driven primarily by sensors and computer vision techniques of matching images and points clouds to specially gathered databases of images and point clouds taken in a range of road conditions? Or will solutions dominate that use GNSS and geospatial databases as the primary reference, augmented by sensors to read exact road conditions, vehicle position and obstructions? How will these systems interface to public and private information infrastructures— for example, to find parking spaces — and to financial systems — for tolls and charges? Will we see a series of closed ecosystems from each manufacturing group or technology provider, or will some interoperability or national infrastructures emerge? Exciting times!
I’m not sure trepidation is the right word, but an area that still requires better definition is that of 3D products. It seems that there are growing market requirements, for example to feed building information modeling or city energy use analysis, but the exact specifications for a profitable, maintainable and national product remain an elusive topic for research and development.