From Medieval Data to Geo-resources on the Web

An Innovative Way of Mapping History

Guido Minini

Department of Civil and Environmental Engineering, Politecnico di Milano, Piazza Leonardo Da Vinci 32, Milano, Italy

guido.minini@polimi.it

1 RESEARCH PROBLEM

This research project is being carried on in the

Geodesy and Geomatics area of a Ph.D program in

Environmental and Infrastructure Engineering. It

raises from a Research Project of National Interest

funded by the Italian Ministry of University and has

the aim to equip Historians with a tool that can

facilitate their work in terms of consulting historical

archives. This project has been dealt with by using a

cognitive approach, in order to identify the key

issues and the appropriate tools to help the

Historians in satisfying their research needs.

In the first phase of the project data collected

from historical sources have been organized into

table archives; these data have been associated with

spatial information (point coordinates),

corresponding to the place names found in historical

sources: knowing the spatial reference it is possible

to import data into a Geographic Information System

(GIS), to display their location on the area of interest

and to overlay them with existing base maps. The

data used for the project date back to the late Middle

Ages (mid-Fifteenth Century) and represent fiscal

information, related to various kinds of taxes paid to

different Kings or Princes in the Kingdom of Naples.

The second phase has been related to the creation

of a relational database to be populated with the

data: as compared with a basic single-table structure,

for example implemented with a standard

spreadsheet, a relational database has the great

advantage to make data query and retrieval much

more efficient, eliminating also unnecessary

redundancies. The relational database design is

achieved through the definition of the Entity-

Relationship Diagram (ERD): in it, data are divided

into different “entities”, corresponding to different

data tables or “relations”, which can be put into

“relationship” among each other, using simple

identification codes.

In the third phase of the project, that has been the

main focus of the Ph.D. program activities, the

problem to be resolved is how to publish these data

in the Web, in order to make them available for the

consultation and query by Historians. Two kinds of

approaches have been explored. The first one is

typical of the WebGIS architecture, in which all the

data and the GIS tools needed to analyse them are

stored in a server and published on the Web, so that

the only client-side software requested is a Web

Browser. The second one is again based on a client-

server architecture, but in this case only the data,

collected into and managed by a Database

Management System (DBMS), are stored in a

server; the GIS tools are provided by a Desktop GIS

installed locally on a PC.

In the next chapters these two architectures will

be described in more detail, comparing them in order

to underline advantages and disadvantages of each

approach.

It is important to state that this research work

presents several challenges, especially from the

point of view of understanding the requirements of

the System’s users. In fact, Medievalist Historians

appear to be usually not well acquainted with GIS

technologies; so defining e.g., a proper conceptual

model fit for XV Century data or implementing

tailored queries represents a major task from the

Geomatics point of view. Several such issues have

been discussed in (Carrion et al., 2015, in press).

2 OUTLINE OF OBJECTIVES

One of the aims of the project is to create a GIS that

can be useful for Historians to draw maps from

historical data in order to study the territorial

dynamics emerging from the historical sources,

achieving a cartographical representation of the

spatial distribution of historical information.

Besides, the possibility to share data among History

research groups is another goal and a feature that

must be added to the system, through the publication

of the Medieval geo resources in the Web.

Currently three historical sources have been

collected into the DBMS:

Minini G..

From Medieval Data to Geo-resources on the Web - An Innovative Way of Mapping History.

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

 the Liber Focorum Regni Neapolis, produced by

the financial offices of Alfonso V of Aragon in

the 1440s, after his conquest of the Neapolitan

kingdom;

 the Quaterni declaracionum, produced between

1446 and 1463 by the leading Treasury officials

(the magistri rationarum) of the last Prince of

Taranto, Giovanni Antonio Orsini del Balzo, the

greatest feudal vassal of the kingdom (Pizzuto,

2009);

 the Quaternus decimarum, drawn up in 1478 by

Paolo Vassallo, bishop of Aversa, that registers

the payments of the Decima, a tax related to

ecclesiastical benefits (Mangia, 2013).

The relational database architecture is useful for

representing this type of data, because it allows to

subdivide the information into smaller parts, making

easier to analyse and query them.

Regarding the GIS technologies to be exploited,

it has been decided to use only open source

software: this goal raises from the need to develop

and maintain the system for a long period with

limited economical resources.

Another important objective pertains the ease of

use of the GIS, because the end users are not GIS

experts, so the system must be complete but also

simple and intuitive to use.

3 STATE OF THE ART

GIS design for historical information analysis is a

rather new achievement, see e.g. (Schlichting, 2008).

At the international level, many works have to be

mentioned, see e.g. (Gregory, 2002), (Gregory et al.,

2002), (Berman, 2005) and (Gregory and Healey,

2007), being the main references for this kind of

research. In Italy some examples of GIS

methodologies applied to the analysis of ancient

maps can be found: (Balletti et al., 2000), (Balletti,

2000), (Baiocchi and Lelo, 2002) and (Balletti,

2006). However the latter works rather address the

urban-historical analysis and the methods for

adapting ancient maps to present-day maps through

geometrical transformations.

Many examples of historical data organized into

national atlases can be found, see e.g (Pawson,

1997) and (Black, 2003); in some cases data are

structured into relational databases, to be managed

into a GIS environment, see e.g. (Ardissone and

Rinaudo, 2005); in most cases they are related to

census data and rarely they date before the XVI

century: (Boonstra et al., 1995), (De Moor and

Wiedemann, 2001), (Fitch and Ruggles, 2003).

It is possible to find some examples of historical

data organized into a GIS and published on the

Internet, such as the China Historical GIS

(http://www.fas.harvard.edu/~chgis/) or the northern

Italy cadastral map WebGIS (Brovelli et al., 2012).

Another interesting example published on the Web

is constituted by the “E 179 database”,

(http://www.nationalarchives.gov.uk/e179/) which

contains records relating to lay and clerical taxation

and which is included into the United Kingdom

National Archives. However, in this case the

geographic component is not made explicit into a

map.

Finally, the research developed and presented in

this paper starts from previous works by (Carrion

and Migliaccio, 2009) and (Barzaghi and Carrion,

2011).

4 METHODOLOGY

This project required a close cooperation between

Geomatics experts and Historians, in particular for

the data collection, data organization and design of

the ERD. Historical know how is fundamental

during the phase of extraction of data from historical

archives and translation of the information into

digital format (typically into a spreadsheet). Also the

conceptual model of the database must be planned

and designed in cooperation with Historians, in

order to understand their needs and the meaning of

the collected information (in particular the

preparation of a data dictionary is necessary).

The participation of Historians in the discussion

about the issues related to the publication of data is

again very fundamental, because they are the end

users of the GIS, so it is important to understand

what tools can be useful for them. Obviously,

technical aspects of the GIS publication concern

only Geomatics experts, and this is the main topic

discussed in this paper.

4.1 Historical WebGIS

The first type of GIS architecture approach adopted

to publish the Medieval data has been an Historical

WebGIS. The WebGIS (mainly developed in the

context of another Ph.D. thesis, see (Zambrano,

2013)) has been the starting point of the research

work presented in this paper. As already said, it is

based only on open source software; its architecture

is shown in Figure 1.

The server side is stored in an Ubuntu Linux

GISTAM2015-DoctoralConsortium

virtual machine and is composed by three modules:

a DBMS, a GIS Server and a Web Server.

The DBMS used is PostgreSQL, an open source

software that can be integrated with a PostGIS

extension, which manages the geographical

information stored in the database and allows to

interact with QGIS Mapserver, a WMS server that

takes advantage of the QGIS libraries and uses the

.qgs projects created with QGIS Desktop. The GIS

server is invoked by the Apache Web Server, whose

task is to publish web pages.

Figure 1: Architecture of the Historical WebGIS.

The client side is entrusted to a Web Browser,

which interprets the web pages written in HTML5

language. The three components of this language are

HTML, CSS and Javascript: they manage contents,

customization and behavior of web pages. The

Javascript toolkit GeoExt was used: it is composed

by the ExtJS library, which allows to improve the

graphical user interface and to integrate them with

grids, buttons and toolbars, and the OpenLayers

library, which allows to include web-mapping

functionalities.

The Historical WebGIS, named Geografie

Medievali (Medieval Geographies) and published

into a Website, is composed by two parts: the

“WebGIS” mode and the “Show table” mode (see

Figure 2 and 3).

However, it must be underlined that the WebGIS

only represents the georeferenced entities of the

database and is based on the architecture just

described, while the “Show Table” page displays all

the entities, georeferenced and not, and some views

that combine the information of two or more entities

in one table, exploiting the relationships between

them; this page is realized again with ExtJS libraries

and allows to query tables and views with a query

builder tool.

Figure 2: The “Geografie Medievali” Website:

“WebGIS” mode.

Figure 3: The “Geografie Medievali” Website:

“Show table” mode.

The typical advantage of this kind of architecture

is that the installation of GIS software is not required

on the client-side, because all the GIS functionalities

are implemented into the server. On the other hand

the GIS tools provided by a GIS server are less

advanced than those provided by a Desktop GIS.

Moreover in the present architecture the “WebGIS”

and the “Show table” pages are not connected to

each other, because they are based on different

versions of the ExtJS libraries. These are the reasons

that led to explore other types of approach.

4.2 Historical Geo-resources Sharing

through the Web

This new solution presents an architecture (shown in

Figure 4) that is much simpler than a WebGIS

architecture: in this case only the GIS database is

stored into the Server; the Client is composed only

of a GIS software, QGIS Desktop, that natively

supports the connection with a PostgreSQL database

through the Internet.

FromMedievalDatatoGeo-resourcesontheWeb-AnInnovativeWayofMappingHistory

This configuration solves the problem related to

the Medieval WebGIS, regarding the connection

between attribute tables and cartographic

representation, which is now implemented.

Furthermore a GIS Desktop software offers many

tools that are not available in a WebGIS

architecture; finally QGIS Desktop is an open source

software that does not require licensing costs and in

recent years is catching up with commercial

software like ArcGIS from the point of view of

completeness and performances.

Figure 4: Client-Server architecture for Medieval

Geo-resources sharing through the Web.

The main problem that remains in this approach

is that the relationships implemented into the

database are not recognized by QGIS, so when the

connection with the DBMS is realized, the entities

imported into QGIS, shown as layers or attribute

tables, are not associated through relationships.

One possible solution is the one implemented the

WebGIS: the creation of views into the database,

that contain the information stored in two or more

entities. The main disadvantage of this solution is

related to the increasing in number of the attributes

contained in a single table, which makes the

consultation of data much more difficult and less

intuitive (if all the relationships in the database

would be implemented, the outcoming view should

become equal to the huge Excel table resulting from

the first phase of the project).

Another solution is being investigated at this

moment, since it could prove to be more

advantageous: namely, to exploit the possibility of

creating some scripts that can work into the QGIS

environment and that allow the software to take into

account the relationships among entities established

in the database.

The entities imported into QGIS from the

Historical Database and one cartographic

representation of the data are shown in Figure 5.

They are overlaid onto a basemap of the Italian

Military Geographic Institute (IGM) made available

as WMS service on the Italian National Geoportal.

Figure 5: The Historical Database imported into QGIS.

5 EXPECTED OUTCOME

The approach presented in this paper has the final

objective to import an Historical Medieval database

into a GIS environment and to implement it in all its

potentiality and completeness (all the entities related

between each other, as represented in the ERD

design). The goal is to supply Historians with an

instrument that makes it possible to realize complex

queries on the Historical data and to show the

outcome of spatial queries on maps.

The System will be accessible mainly by experts

in the field of Medieval research, who will be

provided with suitable tools to easily integrate and

edit data in the database according to their

requirements. The results, in the form of an

Historical Atlas containing selected data and maps,

will be open to a larger community through the

Web.

This type of technology is certainly useful for

improving the ability to deeply analyse Historical

sources and to extract information from them to

support research methods in the field of Medieval

History. Furthermore, even though the GIS will not

be on the Web as a WebGIS, geodata will be stored

on a server and published on a Website, so the data

sharing features of the system will be in any case

maintained with improved performances.

The proposal of this research to exploit the GIS

environment in Historians work can be considered

innovative, particularly for the structuring of a

relational database containing information dating

back to the Medieval period, and for the use of GIS

and Web technologies for History data sharing

through the Internet.

GISTAM2015-DoctoralConsortium

6 STAGE OF THE RESEARCH

Currently the research is focused on finding the best

solution to realize a complete and useful Historical

GIS, from the point of view of the database

capacities. At the moment the final part of the study

is in a preliminary phase, to understand how to

realize scripts to import the database relationships

into QGIS.

In any case the possibility to explore any other

way to solve the problem is not yet precluded, also

considering that GIS technologies are continuously

improving, making new solutions available at a very

fast rate.

REFERENCES

Ardissone, P., Rinaudo F., 2005. A GIS for the

Management of Historical and Archaeological Data. In

Proceeding of the CIPA 2005 XX International

Symposium. Torino.

Baiocchi, V., Lelo, K., 2002. Confronto di cartografie

storiche con cartografie attuali per l’area del centro

storico di Roma. In Atti della VI Conferenza Nazionale

ASITA. Perugia.

Balletti, C., 2000. Analytical and Quantitative Methods for

the Analysis of the Geometrical Content of Historical

Cartography. In International Archives of

Photogrammetry and Remote Sensing, vol. XXXIII,

part B5. Amsterdam.

Balletti, C., 2006. Digital elaborations for cartographic

reconstruction: the territorial transformations of

Venice Harbours in historical maps. e-Perimetron,

Autumn 2006, 1, 4: 274-286.

Balletti, C., Guerra, F., Monti, C., 2000. Analytical

Methods and New Technologies for Geometrical

Analysis and Georeferenced Visualization of

Historical Maps. In International Archives of

Photogrammetry and Remote Sensing, vol. XXXII, part

6W8/1, ISPRS WG VI/3 and IV/3, Bridging the gap.

Lubiana.

Barzaghi, R., Carrion, D., 2011. Il GIS come strumento di

ricerca storica: il contributo della cartografia. Ricerche

storiche, anno XLI, n. 1, gennaio-aprile 2011: 45-60.

Berman, L. M., 2005. Boundaries or Networks in

Historical GIS: Concepts of Measuring Space and

Administrative Geography in Chinese History.

Historical Geography, vol. 33, Special issue:

Emerging trends in Historical GIS. Baton Rouge.

Black, J., 2003. Mapping the Past: Historical Atlases.

Orbis, 47: 277-293.

Boonstra, O., Collenteur, G., van Elderen, B., eds, 1995.

Structures and Contingencies in Computerized

Historical Research. In Proceedings of the IX

International Conference of the Association for

History and Computing. Nijmegen, 1994 (Hilversum,

1995).

Brovelli, M. A., Minghini, M., Giori, G., Beretta, M.,

2012. Web Geoservices and Ancient Cadastral Maps:

The Web C.A.R.T.E. Project. Transactions in GIS 16:

125-142.

Carrion, D., Migliaccio, F., 2009. Il principato di Taranto

in un GIS: problemi e potenzialità. In I domini del

principe di Taranto in età orsiniana (1399-1463).

Somaini, F. and Vetere, B. eds, Congedo Editore: 37-

60.

Carrion, D., Migliaccio, F., Minini, G., Zambrano, C.,

2015, in press. From historical documents to GIS: A

spatial database for medieval fiscal data in Southern

Italy, Historical Methods: A Journal of Quantitative

and Interdisciplinary History.

De Moor, M., Wiedemann, T., 2001. Reconstructing

territorial units and hierarchies: a belgian example.

History & Computing 13: 71-97.

Fitch, C. A., Ruggles, S., 2003. Building the National

Historical Geographic Information System. Historical

Methods: A Journal of Quantitative and

Interdisciplinary History 36: 41-51.

Gregory, I. N., 2002. A Place in History: A Guide to

Using GIS in Historical Research. Oxbow. Oxford.

Gregory, I. N., Bennet, C., Gilham, V. L., Southall, H. R.,

2002. The Great Britain Historical GIS: From Maps to

Changing Human Geography. The Cartographic

Journal.

Gregory, I., Healey, R., 2007. Historical GIS: Structuring,

mapping and analysing geographies of the past.

Progress in Human Geography 31: 638-653.

Mangia, M. A., 2013. La diocesi di Benevento nel

Quaternus decimarum del 1478: un contributo alla

realizzazione di un GIS di cartografia storica sul

Mezzogiorno bassomedievale. Master’s degree thesis,

Università del Salento.

Pawson, E., 1997. The New Zealand Historical Atlas.

Journal of Historical Geography 23: 496-499.

Pizzuto, S., 2009. Il Quaternus declaracionum di

Francesco de Agello (1450-1461). Un contributo allo

studio della geografia politica del Principato di

Taranto in età orsinana. In I domini del principe di

Taranto in età orsiniana (1399-1463). Somaini, F. and

Vetere, B. eds, Congedo Editore: 61-76.

Schlichting, K., 2008. Historical GIS: New Ways of Doing

History. Historical Methods 41: 191-196.

Zambrano, C., 2013. Progettazione e realizzazione del

sistema informativo territoriale “Geografie dell’Italia

Medievale (XIII – XV sec.)”. Ph.D. thesis, Politecnico

di Milano.

FromMedievalDatatoGeo-resourcesontheWeb-AnInnovativeWayofMappingHistory