Case study | Using geospatial data: a guide to licences
This case study examines the research, development and dissemination of our 2019 guide, Using geospatial data: a guide to licences. The guide is a practical advocacy tool that aims to help people understand how they can comply with geospatial data licences and use the data effectively and legally. It outlines, with examples, how data can and cannot be used under different data licences, in various circumstances.
Find out about the ODI's role in developing the guide, the challenges and what we learned.
Summary
The ODI’s ‘Using geospatial data: a guide to licences’ is a practical advocacy tool that aims to help people understand how they can comply with geospatial data licences and use the data effectively and legally.
It outlines, with examples, how data can and cannot be used under different data licences, in various circumstances. The licences covered are used by services such as OpenStreetMap, Ordnance Survey and Google Maps.
The guide is one of the outputs of the wider ODI research and development Geospatial data and technology project which aims to help unlock the value of UK geospatial data – a vital component of a strong data infrastructure. It was developed in response to the ODI’s user research showing that people sometimes found it difficult to understand the language used in data licences, leading to projects being delayed or blocked because of uncertainty and a lack of confidence in how to use the data.
Rather than being a legal document, the guide is an aid to help navigate the complexities of licensing, to encourage reuse of the data.
The use cases outlined in the guide range from creating a map for a newsletter to using data in a journey-planning service to using data in computer games. For each of the 17 use cases, we focused on the licences that users have indicated they sometimes have trouble understanding, and those that will most commonly apply to startups and small and medium-sized enterprises in the UK.
This is the first iteration of the guide, and we plan to build on it in the future, adding other licences and new use cases.
Guide to using geospatial data: key facts and figures
- 17 example use cases, covering different licences and scenarios: from creating maps, to using data in computer games
- The guide includes use cases related to six licences:
- Open Government Licence (created by UK government)
- Data Exploration Licence (created by Ordnance Survey)
- Public Sector Mapping Agreement (created by Ordnance Survey)
- Commercial Licence (created by Ordnance Survey)
- Open Database Licence (used by OpenStreetMap)
- Google Maps
- The guide has been shared with over 13,000 subscribers via The Week in Data, 1 Feb 2019, and shared with 55.5K Twitter followers
- It has over 500 unique page views since launch (Jan 2019)
View from OpenStreetMap
"The guide is an excellent resource for both public and private enterprise. By illustrating different licences using practical examples it makes it much clearer what could be achieved with open data.
"Service providers and consultants should definitely point their clients toward it. It would be great to see this format replicated by licence providers in the future."
-- Jez Nicholson, Director OpenStreetMap UK
What was the ODI’s role?
The catalyst for the guide was the user research – as part of the Geospatial data and technology R&D project – that found there was a need for a user-friendly, plain-English guide to geospatial data licences. The research found that potential geospatial data users were sometimes nervous about using data, partly because they found it difficult to understand the licence wording.
The guide is based around real-world use cases. We worked directly with OpenStreetMap and Ordnance Survey, getting their advice and tapping into their expertise. We did this to ensure that the advice was accurate, to decide which use cases would best meet user needs and to decide which licences to focus on.
Our main goals were for the guide to be accessible and user-friendly, using plain English and clear examples to effectively illustrate the use cases, and show how licences can be used in different scenarios.
When developing the guide, we shared the draft with geospatial data users in our network, using social media. We shared a Google document to gather comments and feedback around the use cases, and the language, structure and usability of the guide. The aim was to incorporate users’ insights and feedback into the document, to help ensure its usability as a practical hands-on guide.
We worked closely with Ordnance Survey, OpenStreetMap other stakeholders to develop and refine the guide. In particular, we worked closely with the Ordnance Survey legal group who answered some of the trickier questions around licences in particular cases. We worked with key stakeholders from early on in the project, to help ensure we enhanced, rather than duplicated, existing guidance, and that we had endorsement directly from the licence providers.
The aim was to encourage use of geospatial data and therefore to maximise the impact of products and services using the data. As well as encouraging use and providing plain-English use cases, the guide includes advice on how to provide attribution when using data under the six licences; and a glossary which defines some common terms used in data licensing, ranging from ‘derived datasets’ to ‘sharealike’ and ‘repeated extraction’.
What was challenging?
The post-project promotion and evaluation of the guide could have been improved. Better planning around dissemination and promotional activities would have helped to increase engagement with the guide, and to deliver impact. Although we marketed the guide via ODI, personal and partner social-media channels, a more targeted dissemination plan may have helped improve reach and impact, and may have encouraged the community to have more ownership of the guide.
The guide could have been broader in scope. There are many geospatial licences available, but for this project, with a limited scope and time period, we chose to only focus on OpenStreetMap, Ordnance Survey and Google Maps. We had initially disregarded Google Maps, as its licensing is restrictive in terms of reuse, but then included it for completeness and to enable commercial comparison. We recognise the small number of licences covered could be seen as a limitation, but plan to address this by adding other licences and new use cases in future iterations.
Earlier engagement with the wider community would have helped with developing the guide. While developing the guidance, the team created an open Google document to gather feedback. This was disseminated through individual social media accounts, with good engagement. A handful of geospatial data users responded and added feedback to the document, and we felt this was well executed given the limited time and staff capacity. But with a more concerted engagement plan we could have engaged differently with the wider user community: in terms of checking the examples, the language and the structure. While this may have been difficult, given the limited time, it may have helped develop a more rounded and comprehensive guide.
We need a more compelling call to action when asking for user input. While there was good engagement via the social-media posts requesting input to the guide (60 retweets and 71 ‘likes’ for the initial post), this did not necessarily lead to conversions: only a handful of people actually commented on the draft document.
It is important to ensure the format meets user needs. The research team, alongside the production team, decided to use the main ODI website for the HTML version of the guide – using a sub-menu and buttons to navigate through the sections. There is also a Google Docs version of the guide, allowing users to download a standalone version of the guide. A PDF version would have also been useful to increase reach and readability for those not familiar with Google Docs. For future, it would be good to build in user testing for the final product, to help ensure quality and uptake, and also around the user experience and usability.
Based on the challenges in this project, for future projects, we should aim to:
- Ensure that impact planning and dissemination activities are costed and built into the project from the start
- Engage earlier – with a more compelling call to action – with the wider community to test assumptions and inform the structure, content and direction of the project
- Build in time to test the final product, and the accessibility/usability of the format and adapt the format according to the research
- Ensure the scope of the project is broad enough, while being realistic about what is achievable with limited time and resource
What went well/what have we learned?
Drawing on expertise at the ODI worked well. We spoke to ODI colleagues to find out more about data licences, what the different terms mean, how licences are used and interpreted, and how to make the guide user-friendly for that community of data users.
The exploration stage is crucial. To write the guide effectively, it was vital that the team understand the data licensing landscape. It was a fairly steep learning curve for members of the team who had not worked with licences before: licensing conditions are complex and the legal language is difficult to interpret. As this was the heart of the problem we aimed to address for end users, it was important to build in enough time for the team to get up to speed. For this project, we had built in the necessary time, which was beneficial in terms of learning and experience.
The cross-organisation working with Ordnance Survey and OpenStreetMap communities has been really positive. They recognised the value in the project – for existing and potentially new users – and were happy to lend their expertise to help answer questions as the project progressed. They were very engaged and our regular meetings really helped us produce a guide that is accessible and meets the needs of the users. As our team are not experts in all areas of geospatial data licensing, we were concerned that we were giving the right advice. It was brilliant to be able to check in with Ordnance Survey and OpenStreetMap when we needed to check a specific clause or condition.
Based on our learnings from this project, for future projects, we should aim to:
- Continue to draw on the expertise of colleagues, members and partners
- Build in time and resource for the exploration stage, to ensure the research team has a solid understanding of the research area
- Work as closely as possible with external experts to ensure relevance and reduce any duplication