By Kari-Pekka Karlsson, The Secretary of the KMTK Project, National Land Survey of Finland. Originally published in Finnish on www.maanmittauslaitos.fi (Read the original version here)
Do we realize how widely spatial data is used in our society? Might it be that not even GIS experts know how often we refer to geographical data, although Google and other widely-used map and spatial data services and products have increased the understanding of the importance of spatial data, or maps at least.
Topographical data, often called reference data, is spatial data in its most basic form. This reference data, typically complemented with other datasets, is used as a basis for most of the operations that are based on spatial data.
Some years ago, I did a small survey on the applications the reference data of the National Land Survey of Finland is used on. I was also surprised by the results. It felt like it’s much easier to list the fields where the reference data is not used.
What Needs to Be Improved and Why?
When the data is used widely, it’s important to put constant effort on ensuring good usability of the data. Furthermore, currently the potential of reference data is not fully utilized – partly because of usability issues caused by incompatible datasets.
National Topographic Database Project Focuses on Development Needs
National Reference Database (KMTK) focuses on the following development need: easy-to-access and relevant basic spatial data is needed to make operations in many areas of the society more efficient. KMTK does not solve all the problems but its mission is to create a basis for a better, more efficient future of spatial data usage that the applications of the future can be based on.
Once I heard a thought-provoking comment from a police officer: “The police car does not start driving unless the navigator is working”. That makes perfect sense. In addition to the security sector, reference data is used 24/7/365 for important purposes in many other applications as well.
There has been a lot of hype surrounding Pokémon Go lately and also speculation around the source of the map data they are using. It should be no surprise that we, map data geeks from Spatineo, just could not stay away from doing our own little investigation of the maps they’re using.
What is Pokémon Go?
Most of you probably have already heard what the game is about, so we’ll just summarize the idea behind the game: Pokémon Go is a mobile augmented reality game where players use their smartphones to catch virtual Pokémon creatures that can be found in real world places. In case you’ve been a little disconnected from the outside world lately and haven’t heard about the game, just check out their video to see how it works.The game relies heavily on maps to not only show the location of the user, but also to show where you can find Pokémon, Gyms, etc. This data covers most of the world and is quite a remarkable data collection in itself, that is why the source of this map data has become a topic of several recent media articles.
And their maps come from…
An articlein The Atlantic was questioning whether the spatial data used in the game comes from the OpenStreetMap or Google Maps, and creators of the game from Niantic declined to disclose this information. So we thought, why wouldn’t we try to figure it out ourselves? We took screenshots of a particular location (in Espoo, Finland) in the game as well as in Google Maps and OpenStreetMap. (Yes, our fellow Finns, while the game isn’t released here yet, we were lucky enough to find a friend who had downloaded the game from a foreign App Store).
Let’s take a look at this screenshot from the Pokemon Go interface:
Just compare the path in Pokemon Go (circle 1) with the screenshots from Google Maps and OpenStreetMap. The features in Google Maps and Pokemon Go are identical whereas the turn appearing on OpenStreetMap is missing from the game.
Google Map Screenshot
Now take a look at the building on this screenshot from the game (circle 2) and compare its’ shape on the Google Map and the OpenStreetMap. OpenStreetMap is clearly more detailed, but all this detail is missing from the game.
These images strongly suggest that Pokémon Go uses Google maps as its basemap. On the other hand, Google Maps outsources most of their map data, which makes it at least theoretically possible that Niantic is just using the same original source instead of Google Maps. However, given the fact that Ingress, Niantic’s original map and location based augmented reality game, was developed under Google’s wing, and that Niantic was spun off only as recently as September 2015, it’s safe to assume they might still have a working relationship when it comes to maps.
Given all this information, we think we can call it: Pokémon Go uses Google Maps. Please let us know if you have information otherwise and we can continue our Pokémon hunt.. er.. investigation.
Why are we so excited about Pokémon Go?
Not only does the game encourage people to get active and explore their surroundings, but it also utilizes map data in a very innovative ways. There is much more to the game’s maps than just the the typical street maps we’re accustomed to. Pokémon Go is backed by a large amount of data gathered from different sources like geology, vegetation and hydrology.
Creators of Pokémon Go have put a lot of effort into using data to give context for the different scenarios the players will encounter. For instance you’ll meet water-type pokémons only next to the actual water features (lakes, ponds, fountains, etc). “That gets into more [geographic information system]-type of data … and we utilise that to map Pokémon species to appropriate habitats,” John Hanke, CEO and founder of Niantic, told Mashable. They might have even used Spatineo Directory as one of their sources, since the Directory has the most comprehensive listing of geographical data in the world and is open for all to browse.
What does a happy marriage between map data and gaming lead to?
Well, so far so good – shares of Nintendo, which owns 33% of the Pokémon company and has a stake in Niantic, rose 53% in the three trading days after the release. Meanwhile both traditional and Social Media is going crazy about the game, as well as most of your friends (probably).
To sum up, what Pokémon Go actually does, in addition to being a highly addictive game, is get people interested in maps and spatial data. This kind of game wouldn’t be possible without up-to-date and highly available public maps. With services like Uber and Tripadvisor, using real-time map data has become a part of our everyday lives without us even realizing it. Now GIS and maps are invading our games as well!
One of the most interesting topics was the status and plans of the INSPIRE Test Framework project. The main goal of this project is to create an open source reference application to verify the compliance of the tested objects against the requirements of the INSPIRE Technical Guidance documents. This work is part of the JRC ARE3NA project and continues until mid 2017. PwC and Interactive Instruments have been contracted to plan and run this software project, which makes me pretty confident about the results.
The main outcome of this work is a reusable, open source software package capable of executing reference implementations of the validation tests for INSPIRE metadata, datasets and services. This software will be hosted by the JRC as a web application, but it will also be possible to install it locally and to integrate the tests in the internal data provision workflows. The project will also define solid data models for the validation testing concepts, enabling persistent test data storage and remote execution and test data exchange using APIs. Needless to say this sounds very promising: A common, well-governed test codebase and a solid software framework is a key in helping the INSPIRE data and service providers to reach the interoperability goals of INSPIRE with considerably less effort.
The initial design of the Test Framework as presented by Clemens Portele of Interactive Instruments at the workshop assumes that the existing ETF tool will be used as the base of the test framework including service testing capabilities using SoapUI and large XML file validation leveraging BaseX. Some of the key extensions for ELF include a properly documented and comprehensive API, multilingual test reports, and test driver for TeamEngine enabling reuse of the existing CITE conformance tests for the OGC standards.
At Spatineo we have been actively involved with the INSPIRE validation work as members of the MIWP-5 group from the beginning. We are very excited to start integrating the Test Framework into Spatineo Monitor to complement our INSPIRE Network Service testing capabilities, which already cover the availability, performance and capacity testing as well as service metadata validation for WMS and WMTS services. I also encourage everyone interesting in this work to take a look at the Design Report of the INSPIRE Test Framework v0.3 containing the initial design and commenting to the authors of something does not look right.
Validate early, provide user-friendly results
At the end of the INSPIRE KEN workshop, Michael Lutz from JRC emphasised the importance of using the validation tools already quite early in the work data provision and service publishing processes: If the validation is only done as the final acceptance test step, this probably results in both extended initial release time and double work as the found errors need to fixed. Other concluding point was the importance of the usability of the validation results: clear error messages, and instructions on how to fixed the particular errors.
No matter how good the INSPIRE validation tools are, using them is not enough to create a successful European Spatial Data Infrastructure. The strength of a good validation tool is guiding the data and service providers to avoid decisions endangering interoperability, and doing that as early as possible in order to save time for more important things.