12th AGILE International Conference on Geographic Information Science 2009 page 1 of 11 Leibniz Universität Hannover, Germany

Description
12th AGILE International Conference on Geographic Information Science 2009 page 1 of 11 Maps and Mash-ups: The National Atlas and Google Earth in a Geodata Infrastructure Barend Köbben and Marc Graham

Please download to get full document.

View again

of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Slides

Publish on:

Views: 8 | Pages: 11

Extension: PDF | Download: 0

Share
Transcript
12th AGILE International Conference on Geographic Information Science 2009 page 1 of 11 Maps and Mash-ups: The National Atlas and Google Earth in a Geodata Infrastructure Barend Köbben and Marc Graham International Institute for Geo-Information Science and Earth Observation (ITC), Enschede, The Netherlands INTRODUCTION This paper is about different worlds, and how we try to unite them. The worlds we talk about are different occurrences or expressions of spatial data in different technological environments: One of them is the world of National Atlases, collections of complex, high quality maps presenting a nation to the geographically interested. The second is the world of Geodata Infrastructures (GDIs), highly organised, standardised and institutionalised large collections of spatial data and services. The third world is that of Virtual Globes, computer applications that bring the Earth as a highly interactive 3D representation to the desktops of the masses. These worlds represent also different areas of expertise. The atlas is where cartographers, with their focus on semiology, usability, graphic communication and aesthetic design, are most at home. Their digital tools are graphic design software and multimedia and visualisation toolkits. The GDIs are typically the domain of the (geo-)information scientists and GIS technicians. Their 'native' technological environment has an IT focus on multi-user databases, distributed services and high-end client software. Virtual globes are produced by hard-core informatics specialists: programmers that specialise in real-time 3D rendering of large data volumes, fast image tiling and texturing, similar to computer gaming technology. Their users however are mostly not aware of, nor interested in the technology behind the products. They don't need that to be able to make sensible use of it (again not unlike computer gamers). We argue that the users of these three worlds are becoming more and more similar. Most of them routinely use computer programs to access them. Nowadays, any standard computer connected to the internet is capable of showing the virtual globes, which not long ago would need the performance of high-end professional workstations. It therefore can also easily be used to view atlas information, and cartographers have indeed been making their information available in mass-consumer environments, first as multimedia presentations on CDs, but nowadays mostly as interactive websites. The web is also the main distribution platform for GDIs, and around the world policy makers have been expressing the intention that GDI information should be available and useable for the general public (the INSPIRE initiative (URL6) is a good example of this). Therefore generally available end-user applications and websites that show or consume spatial data out of a GDI will likely be more and more common in the near future. In the paper we describe the three worlds mentioned, specifically in the Dutch context. We introduce Geo-Webservices as the binding technology between them. The RGI-111 project, whose central goal is the integration of the Dutch National Atlas into the Dutch National Geodata Infrastructure, is briefly introduced. Then we describe in some more detail our tests to further integrate these two with the Google Earth virtual globe. We explain how we could not use actual Dutch National GDI data, because at the time these were not available yet in a useable service form. Instead we took data of similar granularity and complexity, namely the Dutch portion of Open Street Map and served them, mimicking the foreseen Dutch GDI set up. We explain the architecture, using Geoserver to serve the data from a PostGIS spatial database through Network Links and KML to Google Earth. The technical as well as the cartographic optimisations we employed are presented, as well as a discussion of the results. 12th AGILE International Conference on Geographic Information Science 2009 page 2 of 11 THREE DIFFERENT WORLDS The Dutch National Atlas National Atlases are collections of complex, high quality maps presenting a nation. They present a synthesis of the spatial knowledge that characterizes that nation and to that end they contain comprehensive combinations of geographical datasets that completely cover a country. A very important aspect of atlases in general but particularly of national atlases is that all information has been processed to be comparable: the same level of generalization is applied; information has been collected for comparable reference periods; similar classification methods as well as similar class boundaries and legend colours are used. The main point of bringing all the various types of spatial information together in a national atlas is to make sure that this comparison can be fruitful (Kraak et al, 2009; Sieber & Huber, 2007). Added to that is a very important narrative function, i.e. atlases not only present lots of comparable information, but try to combine it into an explanation, a story about the country that gives a more comprehensive understanding than could be obtained from just the separate maps. In other words, there is a synergy where the whole atlas is more than the sum of its parts. A first Dutch national atlas was published by the 'Foundation for the Scientific Atlas of the Netherlands' in the years A second edition followed from 1989 to It was produced by an Atlas Bureau, backed by the Ministry of Education, but when it was completed the Ministry s outlook had changed, claiming that from now on such endeavours as a national atlas should be selfsupporting, and resulting in the atlas bureau closing. Since then the Foundation has been trying to keep the national atlas concept alive, and to this end it was decided to make all maps from the previous two editions available on the web (URL 1). With 1,5M hits since mid 2000 this website seems to answer user needs. But, as the information contained in these maps is getting out of date, new initiatives were needed. (Kraak et al, 2009) Today national atlases are still made as printed books, but cartographers have also been making their information available in digital mass-consumer environments. First as multimedia presentations on CDs, but with the rise of the World Wide Web they soon started developing interactive websites, as early tests for an online Dutch National Atlas demonstrate (see Köbben and Koop, 1997). Until recently, most digital atlases were produced using the traditional digital tools of cartographers: graphic design software and multimedia and visualisation toolkits. And although these result in products that can be of excellent quality and usability in themselves, they are also on themselves: They lack the interoperability and connections with the wider digital world outside that we come to expect in this age of Geodata Infrastructures. The Dutch National Geodata Infrastructure Geodata Infrastructures (GDIs) have been defined by Groot and McLaughlin (2000) as a set of institutional, technical and economical arrangements, to enhance the availability (access and use) for correct, up-to-date, fit-for-purpose and integrated geo-information, timely and at an affordable price, with the goals to support decision making processes related to countries sustainable development. They can be characterised as highly organised, standardised and institutionalised, large collections of spatial data and services. GDI implementation within the European Union is guided by the INSPIRE initiative (URL6). In the Netherlands the INSPIRE implementation is translated into practice guided by the GIDEON report. The ambition for the Dutch GDI, stated in that report is spatial information provision to the digital society (Ministry of Housing etc., 2008). It aims at making spatial information available for a broad range of stakeholders, in a multitude of application areas. The report mentions among many others all levels of government, public services, industry and private citizens. 12th AGILE International Conference on Geographic Information Science 2009 page 3 of 11 The implementation of the GIDEON report and the INSPIRE directive in the Netherlands has been taken up under the direction of the organisation Geonovum (URL12). Many results of these are already available: relevant (meta)data standards and INSPIRE implementation guidelines have been published, a national metadata warehouse (Nationaal GeoRegister) has been established, etcetera. But unfortunately only a few test data services using these were actually 'live' at the time of writing. With such a broad scope, interoperability of the components of any GDI is of course a must. And whereas technical interoperability is already a challenge, semantic interoperability is even more complicated. The producers of the data traditionally worked in isolation, and many of their products were not originally meant for combining with other data sets. The users all have quite different needs, both in the data they require and the use they want to make of that data. This makes using a GDI for any implementation that needs a close integration of its component pieces, such as a National Atlas, especially challenging. A GDI in itself, although it will allow mapping the different data sets, does not constitute an atlas, because the maps cannot be compared and therefore not combined into a narrative. It thus will not constitute more then the sum of its parts, as a good atlas would. Virtual Globes Virtual Globes can be described as computer applications that bring the Earth as a highly interactive 3D representation to the desktops of the masses. Their quick rise to popularity was noted in an editorial in Nature (2006): Millions of people across the world are zooming in from space, flying across continents, and swooping over mountains and through cities, thanks to Google Earth, NASA's World Wind, and other free virtual globes. The most popular of these is without doubt Google Earth. Originally known as EarthViewer, it was developed by a company called Keyhole Inc. EarthViewer became famous during the 2003 invasion of Iraq, when it was used to enhance coverage of the war with a sophisticated use of satellite imagery that was fluid and almost game-like, with its high quality, high frame rate visualisation. In October 2004 Keyhole Inc was acquired by search provider Google, with the intention of expanding search from the non-spatial realm of the internet with a more geographically oriented service (URL8). Google Earth displays the Earth through a combination of different layers of information, including a terrain model that varies from 10m to 90m resolution, imagery from many different sources including aerial photography and satellite images, borders, place names, roads, 3D buildings and links to geo-referenced videos and photos. In addition to these layers, users can load their own spatial data. Many authors have noted the advantages of this platform for scientists and other experts in making their findings more relevant to the general public (e.g. Craglia et al, 2008; Sheppard and Cizek, 2008). Other possible benefits are an increased impact on target communities, enhanced collaboration with other expert users, increased publicity and recognition, and improved fund-raising for research (Sieber, 2006). The appeal of virtual globes for members of the general public seems to lie, as Sheppard and Cizek (2008) write...in the realism, the sensation of flying towards or over the earth, the ability to view their own homes or other meaningful locations, and the sense of their own control over the medium and data: through navigation, manipulation of visible data, and even adding their own data for viewing by others. THE WORLDS CONVERGING As we noted in the introduction, the three worlds we described are converging. We argue this is because the users increasingly access them with similar tools, namely web-based clients and because the underlying technology is more and more based on distributed computing and standardised interfaces, resulting in interoperable geo-webservices. 12th AGILE International Conference on Geographic Information Science 2009 page 4 of 11 The combining technology: Geo-webservices Interoperable geo-webservices are the latest in a long line of software for handling geographic, or spatial, information. In 1962 Roger Tomlinson built the Canada Geographic Information System to determine the land capability for rural Canada by mapping information about soils, agriculture, land use, etc. Tomlinsons GIS software was running on a large mainframe computer and its architecture was what we now call monolithic, with the presentation logic, application logic, and data management layers combined in one software tier. Nowadays larger GISs, like other software systems, have their logical layers separated. Because the client software is separated from the application logic, the same back-end can serve different client platforms. Equally the data being used in the analysis part can come from various different information sources, from various providers, even from different places on the globe. Probably the best-known example of such a distributed system approach is the World Wide Web. This kind of system is only possible when the different components are guaranteed to work together, because they are interoperable. Interoperability in computer systems means in the first place that the systems are able to transfer data seamlessly. This can be achieved by having the data encoded in a standardized, platform and application independent manner. Nowadays, the popular encoding scheme used for that purpose is the extensible Markup Language (XML). Interoperability also means that two information systems should be able to access distributed functionality in a seamless manner. There are many ways of setting up such Service Oriented Architectures, but by far the most used distribution platform is the WWW and the SOA s implemented on that are logically called webservices. If webservices have spatial functionality, for example if they use geographic data, can output maps or find routes, we call them geo-webservices. There are many geo-webservices available, the best-known proponent of it probably being Google Maps. This, and other examples (such as Yahoo Maps, MSN Virtual Earth or MultiMap), can be used by anybody, as their interfaces are publicly available, but they are still proprietary in that they are defined, developed and owned by commercial companies. Alternatively, Open Standards are created in an open, participatory process, where everyone interested can influence the standard. There is a set of such well-defined Open Standards for Geo-Webservices: the Open Web Services (OWS) of the Open Geospatial Consortium (URL3). The National Atlas as a gateway to the GDI Recent technological and organisational developments in the Dutch national GDI have stimulated us to take a fresh look at the national atlas concept. In a recent project within the Dutch national research program Ruimte voor Geoinformatie, a feasibility study and prototype implementation have been executed of automatic visualization of geodata available through the national GDI, in a systematic and cartographically accountable way to guarantee an up-to-date national atlas. The objective was to investigate how the national atlas could be organized as an integral part of the geodata infrastructure. The atlas would benefit from an up-to-date data flow, and the GDI would benefit from integrated visual summaries of available geodata and geoservices in well designed comparable maps using the narrative characteristics of the atlas. As such the national atlas would provide alternative interactive and dynamic access to the GDI (Kraak et al., 2009). The architecture of the national atlas in the national geodata infrastructure will be employing the OWS specifications in a multi-tier set-up. In most existing GDIs the maps are delivered to the enduser from WMSes. Mapping clients combine the output of several such services in a compound map. However, from an atlas viewpoint this is not a desirable solution, as the portrayals of these layers are not matched to each other and to the larger overall atlas goals described earlier. Therefore the National Atlas should primarily use data services instead of portrayal services: It will consume data from GDI WFS sources, and combine and portray that data in a cartographically sound client-side 12th AGILE International Conference on Geographic Information Science 2009 page 5 of 11 application. The data integration and mapping component is used to determine which cartography (map type, classification, etc.) is suitable for specific data layers and their combinations, and which design templates are required to achieve this. To achieve high quality and interactive visualizations, rich vector graphics technology is used to display the maps on the client side. The current prototype, shown in Figure 1, at present implements the mapping client using Flash, the pragmatic reason being we could base that on existing efforts. Having it based on Open Standards, such as SVG, instead of the proprietary Flash technology would in the longer term be desirable. The RGI-111 project and the technical implementation of the prototype are described in more detail in Kraak et al. (2009), and the prototype (in Dutch) can be tested on URL1. Figure 1: Screen dump of the National Atlas prototype (from URL1). Unfortunately, at the time of our project we could not use actual Dutch National GDI services. As explained earlier, there is considerable activity ongoing, but unfortunately only a few test data services using these were actually 'live' and none of these matched the criteria for our test cases, because they were not available yet in a useable Web Features Service form. We therefore are mimicking the foreseen Dutch GDI set up, by taking actual data from the Central Bureau of Statistics (who are planning to be one of the first to have actual WFS available within the GDI), but doing the WFS serving (using the implementation standards already set by INSPIRE and Geonovum) from our own servers. A further step was to see how this could be combined with virtual globes. The integration of these with traditional atlas maps is cartographically rather challenging, as the atlas maps are usually not best suited to be displayed in a 3D environment in combination with satellite or air photo imagery. But technologically it is relatively easy, because software such as Google Earth does allow external graphics to be shown, geo-located within the 'native' layers. To combine the virtual globes with atlas maps in a GDI environment however is a considerable technological challenge. That's because the GDI usually comprises of (very) large data volumes, of varying levels of scale and detail, and the data have to be extracted live from the geo-webservices. Further additions to the cartographic challenge are the limited possibilities for mapping graphics within the virtual globe file formats and APIs. In the 12th AGILE International Conference on Geographic Information Science 2009 page 6 of 11 next section we will describe how we tested out the possibilities by pragmatically putting together existing components and data sources in a manner that is usually called a mash-up. MASHING UP: PUTTING GDI DATA ON A VIRTUAL GLOBE FOR THE NATIONAL ATLAS The term mash-up can be traced back to the first efforts of people to use the then newly released Google Maps for showing their own data, such as Adrian Holovaty's ChicagoCrime.org. Google Maps did not originally have a public application programming interface (API), but developers reverse-engineered it to use the system for their own purposes, and eventually Google released their API (Turner, 2006). Mash-ups can be defined as a web application that combines data from more than one source into a single integrated tool. The term implie
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x