Blog Archives

Feature #04: Introduction to Geographic Information Systems

Your location matters. Despite all the hype about world cities and instant communication, our lives are still constrained and enriched by our immediate surroundings. This blog post is intended to spark some consideration of geographic information in libraries, following on the topic of open access and relating to the topic of visual resources.

Maps are one of the oldest means of writing (abstract representation of things and their relationships with graphic symbols). Archaeological evidence shows a local town plan from Anatolia dated to 6200BC, and many clay tablets from Mesopotamia dating back millennia show other maps of larger regions. Mapping changed as instruments developed and printing presses replaced manuscript copies. The mass market for maps in Europe coincides with the introduction of the printing press.

From Old and New Edinburgh, Vol. III, Ch. 16 http://www.oldandnewedinburgh.co.uk/volume3/page129.html

From Old and New Edinburgh, Vol. III, Ch. 16 http://www.oldandnewedinburgh.co.uk/volume3/page129.html

Originally focused on navigation (by road or ocean), maps expanded in the nineteenth century to present any kind of ‘thematic’ information (such as the economy, population, poverty or disease). The legend around the  1854 cholera maps for London is a common story of the application of spatial analysis (seeing a pattern from the mapped evidence). However John Snow actually proceeded in his analysis, the highly detailed poverty maps of London from the 1880s demonstrate how mapping moved into new realms. Charles Booth and John Snow are now made out to be pioneers of a geographic information technology, when they were much more focused on the content of their maps. In the nineteenth century, as much as the present, maps are a part of science and politics. They present a hypothesis, support an argument, or allow a researcher to examine patterns.

People have been lying with maps long before statistics were invented as a part of modern statecraft. Maps have to lie because they simplify, symbolize, and snapshot a huge complex reality. The hand-drawn map, and its printed successor, are graphic artefacts, and a prime part of the visual resources that libraries manage.  Map libraries are distinctive however because they organize their holdings by location, not author, date of publication, or the regular cataloguing keywords.  Also, geographic location is considered a ‘fact’ by the courts in most countries, making the extension of artistic expression a bit forced. Map makers assert copyright for their choice of symbols, selection of information, and other design components. Due to the rather tight control exerted by Ordnance Survey (OS) on the maps of Great Britain, the Open Street Map movement began in the UK before expanding rapidly worldwide. Issues of open access to geographic information is a huge policy issue with many countries announcing U-turns in recent years (OS now is paid by Government to provide open access).

The contrasting policy of freely disseminated geographic information applies to the US government. A comprehensive street map was developed in the 1980s for the 1990 US Census (called TIGER). Once released for use, the first web mapping applications (such as MapQuest) could emerge without the large cost of digitizing and data entry.

In an era of linked data, geographic location is one of the most common linkages. Recall that Steve Jobs demonstrated a geographic search (nearest Starbucks) when introducing the iPhone in 2007. Of course to pull up Google Maps invokes a huge infrastructure of geographic information that took centuries to put into place.  Geographic references can be used to connect people to the environment, as well as economic flows to the infrastructure that supports them. The basic vision conceives of a series of map layers for each location, each layer derived from a different source, based on different science or administrative rules, but capable of being integrated with the rest.

(Stack diagram, courtesy of Wisconsin Land Information Newsletter 1985, public domain image)

(Stack diagram, courtesy of Wisconsin Land Information Newsletter 1985, public domain image)

The technology termed ‘Geographic Information Systems’ (GIS) emerged over the past few decades, originating in the 1960s with urban and natural resource applications in US and Canada. A GIS organizes various themes as ‘map layers’ in a geographically referenced database. Queries can extend beyond the object itself to nearby objects, either topologically (connection-based) or geometrically (distance-based). All the different themes in one place can be related (as in connecting the pollution in a well to the health of the local residents). Much more complex models can predict floods across a landscape, the course of wildfires, or the dispersion of air pollution. GIS is also used to analyse the pulse of traffic in cities, the delivery of parcels and other services. The breadth of potential applications is huge, spurred by the interdisciplinary content now amassed by GIS providers. Around the world, portals to geographic data resources have been established, following the lead established by the Clinton administration in USA. To a great extent, this represents the Web 1.0 view of dissemination of public data resources. Initially the data providers were national mapping agencies (such as Ordnance Survey in UK, US Geological Survey and Geosciences Australia). The content of these portals is now subject to ISO standards for structure and metadata. Recently the EU programme INSPIRE has required additional harmonization for EU members, a project that has cost millions of pounds.

Alongside the national portals, the larger bulk of data is held in more dispersed hands, mostly local governments but also private corporations. Local authorities work at the finest scale, with detailed information to support permits, taxes, and day-to-day operations. They maintain equipment and installations like lamp-posts, drains, and unglamourous necessities of modern life. Each of these objects gets described with a coordinate in a geographic database. In industry, some require geographic search to manage resources such as forestry, agriculture or mining. Other industries use GIS to reduce costs, often by modelling logistics.

GIS users also consume vast amounts of imagery and sensor data from air photographs and satellite images. These images are visible as a background in Google Maps and their competitors. Forty years ago, the first experimental remote sensing satellites were launched, providing images with pixels of 80 meters. Now the commercial satellites can provide 2 meter resolution or finer. The difference is immense, since there are 1600 times more pixels (in the switch from 80 m to 2 m). Computer storage has become cheaper and more plentiful, but we manage to fill it up. More and more nations have invested in earth observation (EO) satellites, once the preserve of just a handful of countries. Last month, Vietnam joined the list (with a satellite built by French industry with substantial subsidy from the French government). In the next few years the number of EO satellites will more than double. In addition to satellite data sources, air photography has remained important for its ability to fly under clouds and at higher resolution in urban areas. Recently, terrestrial camera systems have become popular with the uptake of services like Google StreetView. These systems can also carry laser distance measurement devices (called LiDAR) that can measure millimeter accuracy with gigabytes of data on urban streetscapes. As more and more sensors are available, the need for GIS increases just to manage the sources.

GIS software developed in the 1970s, largely designed for specialists. The market remains a niche one, with a few big players (notably Esri from California, Bentley, Intergraph, and some competition from the open source sector) licensing packages for substantial annual fees. Access to geographic analysis for the mass market has become a key element of the large web corporations, notably Google. The model is one of services, such as the shortest path provided by Google Maps between two addresses.

More recently, the flow of data has become two-directional. Web users contribute their geo-tagged information to build data resources that compete with the top-down traditional providers. Open Street Map is often quoted, but there are specialist groups around the world building resources for plants, animals, or whatever interests them. Despite fears that non-authoritative data sources might confuse more than they inform, the data quality of crowd-sourced information has proven to be at least as high as official mapping in certain (more urbanized) areas.

How did this happen? First of all, the computer, which used to be hard to access, is now more universally available. More specifically, of course, geographic positioning has become much cheaper and widespread as well. Surveying technology used to be time-consuming and meticulous, often taking days for a large crew just to extend the measurement network by a few kilometers. Now a hand-held device can measure quite accurately in seconds due to the Global Positioning System (GPS). Originally developed for military applications, GPS has become a term for not just the measurement of location, but the whole suite of databases and visualization tools that accompany the device. The consequence is that every mobile device has the capacity to receive geographic information about its surroundings, and to contribute additional details or corrections. Current users expect the map to be built around them, rather than the tedious process of locating yourself on the map.

It is hard to sum up a whole field in a few words, but here is my definition of GIS:

“Geographic Information System (GIS) – The organized activity by which people

•             measure aspects of geographic phenomena and processes;

•             represent these measurements, usually in the form of a computer database, to emphasize spatial themes, entities, and relationships;

•             operate upon these representations to produce more measurements and to discover new relationships by integrating disparate sources; and

•             transform these representations to conform to other frameworks of entities and relationships.

These activities reflect the larger context (institutions and cultures) in which these people carry out their work. In turn, the GIS may influence these structures.” (Chrisman 1997, Exploring GIS, New York: Wiley and Sons, p. 5)

(Ring diagram, from Exploring GIS, Wiley and Sons, by permission)

(Ring diagram, from Exploring GIS, Wiley and Sons, by permission)

The key trick is converting from one view to another; taking points and interpolating a surface or bringing environmental data into an urban analysis. The field is prospering, delivering results in many applications.  One recent study estimates a huge impact on the world economy with value added of about $100 billion per year.

(illustration: http://www.oxera.com/Oxera/media/Oxera/images/Oxera-Geo-Services.jpg )

Where do librarians fit into this current era? In some research institutions, the map librarian has become a GIS specialist, curating a collection of digital data, and providing access to tools (software).  Some work on digital libraries focused on using the locational element to index all library holdings. For example, connecting novels to the places they describe, and tracking artists across the landscapes they painted. Still, most GIS these days remains in specialist hands, used to manage cities or reduce cost in logistics for enterprises.

Nick Chrisman has been working in the field of geographic information for 41 years, developing innovative techniques to analyse and display information about the earth and the people who inhabit it. He is currently Head of Geospatial Sciences at RMIT University in Melbourne, Australia. As a programmer at Harvard’s Computer Graphics Lab in the 1970s, he was part of a team which developed early computerised mapping systems – what we now call GIS (Geographic Information Systems). He has worked as a professor at the University of Wisconsin, the University of Washington, and the Université Laval. (ca.linkedin.com/pub/nick-chrisman/5/977/a03/ )

Some resources:

Old Maps Online (shares digitized images from many countries) http://project.oldmapsonline.org/collections

British Library (scanned collection of great depth) http://www.bl.uk/reshelp/findhelprestype/webres/scanned/

David Rumsey’s map collection (mostly North American, but some other regions, very well presented) http://www.davidrumsey.com/

Vision of Britain (1801-2001); download for UK Federation universities only. http://www.visionofbritain.org.uk/maps/

What is GIS? (hosted by ESRI) http://www.esri.com/what-is-gis

Open Source Geo http://www.osgeo.org/

Economic value of Geo Services (Oxera commissioned by Google) http://www.oxera.com/Publications/Reports/2013/What-is-the-economic-impact-of-Geo-services-.aspx

#uklibchat – the Prezi

In July 2012,  I gave a presentation on #uklibchat for ARCLIB‘s 25th anniversary conference:  Shaping the Future

The following Prezi is a slightly different version of the one presented. If you want find out how and why we got started, who our users are, and what they like about #uklibchat, please check it out!

Ka-Ming

A Review of #UKLibChat’s First Session

The first #UKLibChat discussion took place last week, and we are pleased to say the event featured on Information Today Europe‘s website, entitled “Twitter Chat: Instant ideas and collaboration“. The article outlines the history of Twitter Chats, what you can gain from participating and how to get involved. The full article can be found at the following stable URL:

http://www.infotoday.eu/Articles/Editorial/Featured-Articles/Twitter-Chat-instant-ideas-and-collaboration-76503.aspx

Happy Reading!

Follow

Get every new post delivered to your Inbox.

Join 2,511 other followers

%d bloggers like this: