1From a certain point of view, cartography and Web cartography seem to be the antithesis of each other. The first discipline is over a hundred years old and deeply connected to its traditional roots. The latter is a very modern Internet information service, characterized by a continual and vertiginous evolution. However, this initially seeming paradox has turned into a truly collaborative relationship with the arrival of the map digital format and Geographical Information Systems (GIS). The Web has changed our outlook on the role of maps (Kraak, 2004). The map - traditionally an abstraction of geographic reality - has gained some new functions in the Web context, connected with its new role within a web search engine. A digital map may contain text or graphics linked to the geographic elements or alternatively links to other maps on the Internet. In addition, the map itself can act as a research tool within a Web search engine (for instance, when researching a certain topic using a territorial base). Finally, the map can be a link, guiding the Internet user when downloading certain geographical data.
2Moreover it is important to bear in mind that the digital format of cartography has been recently and efficiently disseminated by the Internet. This is demonstrated by the impressive number of software tools which can draw digital maps at very low or even no cost. Most of these tools are a sort of light version of the more complex GIS, which as is well known, can be used to draw maps amongst its other functions. The drawn maps are in fact only one of several ways to communicate spatial analysis results, on the basis of the queried data from the geographical database connected to the GIS. Quite often the media quote and use maps in order to clarify the news they publish. People now seem to appreciate knowing where the action - related by newspapers, television, advertisements etc - takes place. But this renewed interest in Geography and Cartography, which has been positively valued by the interested categories, has been made possible by the development in several interconnected sectors, for instance computing. Software versions are updates their versions very quickly in order to continuously improve their functionality and gain a corresponding exponential growth of their producer’s gains. On the other hand, the Internet renews itself and its related services and converts an increasing number of its users into collaborators.
- 1 The term World Wide Web was officially born on the 13th March 1989, with the publication of Tim Ber (...)
3After twenty years the Web has become participatory and has gained a brand new inscription (Web 2.0), which reflects the evolution under way.1 As already briefly pointed out, digital cartography is also hosted inside the Internet and propagated by it. We can identify two main typologies of Internet mapping sites, which are characterized by ownership or non-ownership of their cartographic bases.
Websites with their own cartographic base (or a map base supplied, more or less for free, by an outside institution as a numeric regional map in a vector format or an orthophotograph in a raster format). On this basis, several thematic layers are then overlaid, drawn by the website owner. An instance of this kind of website could be those of University Department, which publishes the results of its geographical studies as thematic layers in a vector format. These layers are overlaid onto a numeric regional map and can be queried using a mouse in a graphic environment, in order to gain their text attributes, which are recorded on database tables. The numeric regional map has been supplied by the public institution that produced it.
Websites with their own thematic layers overlaid onto an external cartographic base, which is stored on remote host servers on the Internet. The cartographic base has been obtained through the API (Application Programming Interface) method.
4This paper, following a brief introduction on some technical options connected to the two different kinds of websites cited above, will critically assess the cartography employed in the two different instances. With regard to this last aspect, in the third paragraph there are some remarks on the inappropriate use of the digital cartography in Internet. The insufficient knowledge of mapping basic rules can cause an improper use of the GIS software mapping tools in the Web. The result can be an imprecise or even wrong map, both from a formal and (or) from a substantial point of view.
5There are several different Web mapping, WebGIS and Internet GIS definitions one can find in specialized bibliography. The truth is that they are all often overlapping concepts, because the thin line which separates them is only visible if we consider their functional differences. In other words, we can avoid the troublesome confusion if we question the main feature of the single tool under consideration. This can be easily deduced from the literal meaning of the words composing its name.
We begin with the last term because we consider it far more extensive than the previous two. Some Authors consider Internet GIS mainly as a framework of “network-based GIS that uses the Internet to access remote geographic information and geoprocessing tools” (Peng et alii, 2003). Others highlight the interface, the application for processing and the database as the different elements that compose the so called Internet GIS client-server structure (Penev, 2006). Often the interface element is a map server (called WebGIS), with the task of drawing the maps that are queried by the Web browser clients, on the basis of the spatial analyses performed by the GIS. In conclusion, Internet GIS can be considered a GIS produced spatial analysis diffusion using the Internet.
- 2 This software can be a common html page displayer (for instance Mozilla Firefox), a more complex ge (...)
6The generic term “Web mapping” can be applied to a procedure aimed at creating and distributing cartography using one or more Internet host servers for a diffuse client computer end user. Of course, the client computers are connected to the Internet and are often merely equipped with simple software such as a browser.2Nowadays there is a lot of talk about Web mapping 2.0, in order to underline the switch from a static map, stored on the host server, to a dynamic one, built on the base of a remote client query (Gartner, 2009).
- 3 Web mapping service identifies, instead, the possible downloading of maps from the net.
On the otherhand WebGIS, a very close Internet GIS relative, more often refers to the drawing functionalities, which can produce maps as a complex or simple spatial analysis result, from the basis of an Internet user query. The queries are often performed by a simple browser in the World Wide Web, the widespread Internet information service working environment. The WebGIS functionalities can be applied to a base map stored on one or more interconnected servers, also managed by different institutions, which can integrate and share their databases. Here we could cite the Web mapping Server3, in reference to the geographical information distribution on the Net through a heterogeneous and widespread source.
7WebGIS can distribute maps either in a dynamic or in a static way. In the former case, the users require digital maps in an image format (jpg, tif, gif, etc), but do not have the opportunity to interactively customise the maps (i.e. requesting particular thematic features, geographical area, scale, etc.). In the latter, the users can raise queries with the database. These queries trigger some data procedures by the WebGIS in order to draw the different maps queried by the users. In this case the system delivers a true spatial analysis, which can differ in complexity, according to the implemented functionalities. This can lead to the creation of different maps, customised by the different provisions imposed by the users.
It is not my aim to examine methodologies in depth or analyse the problems connected with making and downloading maps. The bibliography references a couple of texts, which could be useful in relation to these topics (Peng & Tsou, 2003; Dragicevic, 2004). The importance lies in pointing out that regardless of the geographical model or software adopted, in every WebGIS structure the cartographic base is managed entirely by the Website administrator. In other words, the original vector geographical data published in the WebGIS are overlaid on an ad hoc base map produced or simply leased from a third contractor.
8A low cost alternative to creating a WebGIS is to opt for an external base map by exploiting an external software development environment, which comes free with the maps. This is called the API (Application Programming Interface) development environment.
- 4 A Mash-up site is a hybrid Web application, which includes information from different sources in a (...)
9The API is a very popular tool in the computing field, used initially not for cartographic purposes well before the arrival of the Mash-up sites on the Internet.4 The API is a set of computing procedures used by the programmer to expand software functionalities in order to reach a certain aim. It can be considered as an abstraction between low and high level software and it enables the customisation and improvement of software functionalities without having to write all the starting functions of the software itself. Such API application can be seen on the website of a University Faculty, which shows the location of the academic headquarters on Google maps, instead of producing a static map by itself. On the University website all the dynamic functionalities of Google maps are allowed, as enabled by API.
Figure 1. API application instance: Faculty of Education Sciences (Trieste University), location of head office and branches on the Google Maps map base
Please see in the upper right part of the figure the given choice of pinpoint the Faculty on a Map, a Satellite scene or a so called Hybrid version. This last overlays the street network and some text information on the satellite scene.
Source: http://www.units.it/strutture/index.php/from/didattica/area/didattica/menu/didattica/strutture/009000 (last access: March, 21th 2011).
10There are several non cartographic examples of API application, which include the PC graphics interface connection to BIOS (Basic Input Output System) and Windows API.
BIOS enables access to a schematic environment and is easy to understand and use. This can be useful to set and control some basic functionalities of the hardware and of the principal input and output device management.
Windows API is free from Microsoft together with the operating system. They permit improvement and personalization of Windows in the graphic field (2D and 3D), in the memory management, the game device features (for instance the joystick), the sound card, etc.
11Generally, API is distributed in order to increase software diffusion. In the case of Microsoft , it provides Windows API (in this case called open API) for free, whilst restricting the use of other API (closed API in this case) only to registered software programmers, for instance the ones connected to the development computer games for the X-Box console. API is very widespread on the Internet due to the fact that it is relatively easy to use. This tool simplicity (or complexity) depends of course on the final aim of its users. In any case, we would like to reinforce the concept that this tool encouraged (and still does) full user participation, many of whom employ API to build Web applications.
- 5 A certification Authority (CA) is an official entity that issues digital certificates, which certif (...)
12The recent Web evolution finds its users assuming leading roles with respect to the appearance and functions of the Internet. The case in point is Web 2.0, which underlines the evolution under way and some analysts initiated soon after the Internet market bubble burst, between the end of 2001 and the beginning of 2002. During the so-called “new economy” or even “dot-com” period throughout the second half of the ‘90s - several Internet-oriented firms boasted a very high profit share. In this period several companies emerged, including Google (1995), the celebrated Internet search engine, and the Certification Authority Thawte (1995).5 Web 2.0 essentially implies participation, Internet users are no longer merely consumers of the information stored on the various Internet host computers; they are now true contributors. There are several social networks on the Internet operating along these lines including the free participation encyclopaedia Wikipedia, which lists as many as 691.000 different entries in the Italian version (25th May 2010). Another example is Flickr, a website where anyone can publish their photographs, which can be used by others or finally Ebay. The latter popular virtual point of sale numbers 89 million users in 39 world markets.
- 6 Amazon decided to open itself to the cloud computing market in 2006, offering various computing ser (...)
13From the company point of view, Web 2.0 is a platform which supplies services in return for payment via the global phenomenon known as Cloud computing. The cloud is a metaphor for the Internet, which offers services to firms and people, who pay for what they use (the so-called “pay per use”). In this way the firms obtain economies of scale, sparing the fixed costs connected to the use of software. We no longer need to buy and update expensive software, which requires complex installation and validation procedures due to the spread of global hacking. There is also no longer a need for expensive servers, vulnerable to foreign intrusions. All you need is a PC with a standard hardware configuration, connected online in order to use the cloud software services, which are outsourced and thus available everywhere in the world. This is a huge geographical shift in terms of global computing procedures. Cloud computing already exists and in the not-so-distant future ay trigger a real revolution in the IT field, of similar consequence to the one spurred by the DOS advent in 1982. Some examples include Google docs, the free writing platform from Google; Amazon Web Services, the computing services offered by the famous Seattle firm, traditionally specialized in e-commerce;6 the simple availability of a mailbox; the rental of a web space c/o a host computer provider.
- 7 Craigslist is an online virtual classified advertisement newspaper. The advertisements can be of va (...)
14The exchanges between Cloud computing, Web 2.0 and API are behind the creation of Mash-up sites, which draw on information from several sources to create a derived product. A mash-up site is a website which duplicates information on the net with a reconfigured content and a new appearance. Mash-up sites also involve cartography. Many appeared on the net, for instance, using the base maps and functionalities of Google Maps, joined to different kinds of information linked to the maps. We can take the Mash-up site in Fig. 2 as an example. It is a combination of real estate advertisements on Craigslist7 with Google maps: the result is a very efficient aid for real estate sales. We can see another instance in Fig. 3, which shows where we would emerge on the opposite side of the Earth if we kept digging starting at the centre of Trieste, Northern Italy.
Figure 2. Cartographic Mash-up website, which connects the real estate advertisements of Craiglist to the Google Maps base map
See the left part of the figure. In the upper part you can see the house location on a Map, on a satellite scene or on a relief map, a sort of Digital Terrain Model with the principal streets and some information overlaid). In the lower part you can see the house for sale photograph (Birdseye, Streetview or WalkScore mode. This last gives some information about the closest shops, clubs and restaurants).
Source: http://sfbay.cribq.com/, last access on March, 22th 2011.
Figure 3. Cartographic Mash-up website showing where we would emerge on the opposite side of the Earth if we kept digging starting at the centre of Trieste, in Northern Italy
You can see the digging pinpoint (centre of Trieste town, Northern Italy, coordinates: 45°37′46″N, 013°46′37″E).
Source: http://map.talleye.com/bighole.php, last access on March, 22th 2011.
Figure 4. Cartographic Mash-up website showing where we would emerge on the opposite side of the Earth if we kept digging starting at the centre of Trieste, in Northern Italy
The end up point (Pacific ocean, East of New Zeland, coordinates: 45°37′46″S, 166°13′23″W).
Source: http://map.talleye.com/bighole.php, last access on March, 22th 2011.
15Before remarking on the cartographic conformity of maps produced through WebGIS and Mash-up sites, it is necessary to point out the substantial differences which exist between any tool and its subsequent output. Any instrument should be neutral regardless to its production. If we apply this principle, when examining GIS and digital mapping we can observe that some distinctions are appropriate, especially in relation to the tool functionalities and the features of its production. There is no doubt whatsoever that GIS is not simply a producer of digital maps but can also prepare cartography using the space analysis by which it is characterised in the technical/scientific field. According to Bianchin (2009), the relationship between cartography and GIS objects can be understood if we consider the third principal GIS functionality after the spatial and the graphic analysis, that is to say data management.
- 8 Nowadays in Italy the majority of geographical digital data is in the shape of regional numerical c (...)
- 9 Referring to the graphic error and to the connected symbol density on a map, as is well known, the (...)
- 10 Bianchin, on the subject, maintains that: “a map scale is not only the metric relation between the (...)
16A GIS is also, in fact, a geographical data deposit. In the cartographic field geodatabases or topographic databases are mentioned quite often. Here the spatial objects that can be used to draw a map8 are memorized and can be easily accessed. The topographic database relates to a certain scale that, by means of a different imposition of area length ratio (map and reality), provides a key to the land representation (and to the land interpretation as well). The different objects registered in a topographic database may also form a cartographic database on the same scale, which “takes into account the symbols’ overlying problems and solves those following cartographic rules in order to reach a good cartographic resolution” (Bianchin, op. cit.). From the cartographic database we can finally obtain the map (cfr., for instance, Laurini, 1992 and, in order to deepen the so called “Chorem” visualisation of the Database extracted information: Brunet, 1986; Laurini et alii, 2006; Laurini, 2009). Using cartographic generalisation procedures we can later derive minor scale maps (for further study with particular regard to the changeover from manual to automatic generalisation, see Bianchin et al, 2003). As Amadio suggests (op. cit.), GIS’s handling of geographic information in a vector digital form, combined with its storage capacity through its databases overcomes the tie between the preponderance of graphic elements and the object density on the map.9 Theoretically it is hoped that the survey can be carried out at a bigger scale along with the topographic database built on this base. Using queries aimed at the database, the objects of interest could be filtered, in order to compose a map on a minor scale, derived from the topographic database. For practical reasons, Amadio specifies that the standard procedure builds several topographic databases at various synthesis levels, derived from the surveyed database. Pending the completion of the IGM topographic database, in Italy the regional numeric cartographies available are often downloadable at no cost from the regional websites. Since they are free, regional numeric maps are often used as a cartographic base in several WebGIS, upon the map source quotation. On this base map the thematic maps realised by the WebGIS are then overlapped. Generally speaking, regional numeric maps can be considered semi-finished goods, because they are composed of a lot of layers (geographic elements in the shape of vector points, lines and polygons), sometimes so thick that the map is difficult to read. They are halfway between a topographic database and a map and so they need further data processing in order to lighten and customise them (for instance by the colour and the thickness of the symbols), before being used. Meanwhile the overlapping vector layers are often not very homogeneous with the rest of the map because of their thematic characters, symbols used, and in the worst cases, also due to their scale.10
17What follows are a few examples of the most clearly visible cartographic mistakes that can often be noticed on WebGIS mapping displays (it should be noted that they can also all be found on GIS-made digital maps):
maps with unsuitable symbols because of their typology and dimensions. Symbol dimensions can particularly be out of shape at certain zoom levels, they overlap each other and subsequently hide the underlying symbols.
- 11 Generally, in a GIS environment, the zooming of the display has a connected reference scale. Allowi (...)
the symbols are not linked to the real geographical object dimension; there are too many graphic errors in a GIS environment, due to the facilitated zooming capabilities of the display;11
in the thematic map many different colours have also been used to differentiate the geographic elements, causing confusion and making it impossible to link each object to its thematic class;
topographic maps, which therefore represent small areas, use an unsuitable kind of projection, because this has been developed for thematic mapping at a minor scale.
- 12 See the passage: “Del rigore della scienza” in the book “L’Artefice”, Italian translation, edited b (...)
18In order to counter the problems resulting from the thickness, density and compliance of cartographic symbols, the Internet proposes the use of a remote sensed image as a base for any vector layer, linked to photographs, drawings and text, as in a Web hypertext using geobrowsers and Mash-up sites. There are those who strongly affirm that the famous Borges dream of the 1:112 scale map, has been realised with the aid of Google Earth/Maps, the most complex map ever created (Bompan, 2009).
- 13 For further details on geobrowsers and Website cartography see also Favretto, 2009 (a, b).
The problem is that all this is not cartography; rather an interactive land visualization on the computer display. Bianchin (2009) observes “the object topology on the ground is complete, but there are no objects at all, only a photographic continuum” (2009). As many Authors already wrote (cfr., for instance, Robinson, 1976 and 1995; Bertin, 2010; Brunet, 1987; Mac Eachren, 1995; Harley, 2002), a map is essentially a territorial representation, not the territory itself (captured by an aerial photograph or at a hypothetical, absurd 1:1 scale). Not all the geographic objects can be positioned on the map and this in any eventuality would certainly not be desirable. The result would be an unclear map. In order to better understand this, we can look at an aerial photograph of a known urban area using, for instance, Google Maps. Even if we know the territory represented in the aerial photograph, we might not be able to recognize the locations, so we need to use a so called “hybrid visualization”. This is the overlay of a vector format road network on the photograph, with the addition of some street names. Of course, the same problem may arise in the case of suburban areas. It seems that on Mash-up sites everything is permissible in terms of contributions overlapped onto the remote sensed images. The aim is to enrich the bird’s eye view of the image with information. The fact that the tool is free means its users accept any mistake or inaccuracy. Let us consider as the example of Wikimapia, the mapping website where anyone can insert vector elements on a Google base map, imported by the API. The display appears thick with various shape polygons (Fig. 4), corresponding to the many advertisements/information added by Internet users. The advertisers therefore gain a free publicity channel online; Google supplies the maps (acquired from Tele Atlas), covering the cost through free advertising of its search engine. All this seems to be an efficient business rather than digital mapping.13
Figure 5. Wikimapia screen showing Trieste and its surroundings
On the satellite scene are clearly visible the users drawn polygons that highlight a certain place or a tourist attraction or a business. Clicking with the mouse a drawn polygon a window opens with text information, internet links and photographs. In the lower part of the figure is shown the information window concerning the Rossetti theatre of Trieste. Please see the internet links and the information on the left side of the window while on the right one there are some advertisements.
Source: http://wikimapia.org/#lat=45.6486&lon=13.78&z=10&l=4&m=b, last access on March, 22th 2011.
19The summary concludes - in light of the facts abovementioned - that it is futile to question if WebGIS can produce cartography or not. Most of the time WebGIS can produce good territorial views, but they can be rarely considered true cartography.
GIS should instead, be considered as formidable territorial analyst. Through the implemented database technology, GIS can store all geographical data and update it via automatic procedures. This results in maps which are at the same scale as the topographic database or at a minor scale (thanks to the automatic cartographic generalisation procedures).
20In the future, information technology will be increasingly present in cartography at all levels of data elaboration and also at a mapping level. The printing option of the map digital format will either be applied to the whole map or to a part of it. GIS should be considered as a valuable cartographic partner. The knowledge of GIS methods will thus substantially enrich the cultural and technical background of the cartographer.