@misc{Constantopoulos2019b,
title = {From publications to knowledge graphs},
author = {Panos Constantopoulos and Vayianos Pertsas},
year = {2019},
date = {2019-11-12},
howpublished = {Talk at World Health Organization Global Technical Meeting 2019, Seoul, 12-14 Nov.},
keywords = {},
pubstate = {published},
tppubtype = {presentation}
}

@misc{Constantopoulos2019,
title = {From the catalogue to the virtual information space},
author = {Panos Constantopoulos},
year = {2019},
date = {2019-08-24},
abstract = {Libraries have always been knowledge vaults, serving as indispensable infrastructure for research, education, and an array of important social functions. With time, sophisticated knowledge organization systems have been developed that support the access to the content of libraries. Effectively, these systems reflect our perception of various domains of knowledge through agreed, yet evolving, organizational schemes. The contents of libraries also undergo an important evolution: not only do the documents of knowledge become increasingly digital, they also come at widely different media and levels of granularity, from books, to articles, to images, tables, datasets, video, audio, etc., each independently identified. In addition, the entanglement of research processes with information processes becomes tighter in digital environments. In this talk we will review, in the context of these trends, the potential for knowledge access and integration offered by ontology-driven semantic graph indexing. We will also try to show that a wider margin for effective knowledge access is enjoyed when data- and process- oriented approaches are combined, especially in view of the increasing ability to use automatic knowledge extraction and indexing techniques. Libraries are thus facing the opportunity and challenge to create connected information spaces rendering, as close as possible, the rapidly evolving body of knowledge.},
howpublished = {Invited talk at IFLA World Library and Information Congress, Athens, 24-30 Aug.},
keywords = {},
pubstate = {published},
tppubtype = {presentation}
}

@misc{Ilvanidou2019,
title = {And The First One Now Will Later Be Last, For The Times They Are A-changin': Modeling Land Communication In Roman Crete},
author = {Maria Ilvanidou},
url = {https://dev.clariah.nl/files/dh2019/boa/0632.html},
year = {2019},
date = {2019-07-11},
abstract = {The present contribution has a twofold aim: on the one hand it will seek to demonstrate how the use of digital tools and methods enabled the reconstruction of the road network in Crete, Greece during the Roman period (1st century BCE - 4th century CE), while on the other hand it will showcase how the rapid developments in digital tools often deems research in the field of the Humanities outdated or obsolete. Back in 2005, when I first started working on digitally modeling land communication in Roman Crete, the puzzle I was trying to put together was looking for the bits and pieces of relevant and useful information within a variety of diverse and scattered sources: ancient written sources (like, for example, Strabo’s Geography and Stadiasmus Maris Magni , an ancient Roman periplus detailing the ports on the shores of the Mediterranean Sea), archaeological evidence (paved road remains, bridges, miliaria, i.e. Roman milestones, and Roman sites), historical maps (like, for example, the Tabula Peutingeriana , a medieval copy of an itinerarium pictum, i.e. a painted itinerary, showing the layout of the road network of the Roman Empire, and Venetian maps of Crete), travel literature (15th-20th century travelers’ accounts) and topography (largely my own surveys).

Material heterogeneity and diversity of information and data collected called for sophisticated modeling; that is, an information schema that would capture and document any determinant detail, and, when implemented, be able to facilitate correlation and answer my research questions: which settlement types (for example cities, sanctuaries, farmsteads) were connected through the road network, what were the distances between them, how long it took to travel and by what means of transportation (e.g., by foot, horseback, mules, etc.), which route was followed by which traveler and for what purpose, do the routes mentioned by different travelers change across time, how accurate or credible is the information provided by ancient sources with regard to distances between settlements, what parameters affected the course of the route and the planning of the road, to what extend Cretan topography determined the route direction, what were the local topographic characteristics that affected transportation on the island, and can we reconstruct the original trace of the Roman road network with the use of digital tools and methods even though the archaeological evidence is scarce and fragmentary? The other challenge was to combine the use of Geographical Information Systems in order to restore visually the spatial data and exploit the GIS functionality to check and assess parameters that affected the planning of the road network in question.

The digital tools that made this endeavor feasible were two. The first one was ArcGIS (ArcView 9.0), a commercial GIS software still largely in use, for the spatial analysis of geographic data. The selection of the second one, the tool that would enable me to manage, store and correlate all historical and archaeological data, has proved more challenging and changed over the years. From MS Access, for the implementation of a relational database, back in 2005, to BetaCMS, an open source web-based content management platform, which used XML schema definitions to represent content, in 2010. The BetaCMS, later called Astroboa, was developed by a Greek IT company, Beta Concept, and allowed fast and easy modeling, storing and querying of all data, thus providing what it seemed to be, at the time (2010), a suitable alternative to switch into. The alternate and combined use of both systems (GIS and database) allowed for a long and intriguing iterative process from which a network of optimal paths emerged as a result. This enabled me to propose a reconstruction of the original trace of the public road network connecting the major cities and settlements of Roman Crete, and in specific cases test it against field trip data, with very promising results.

Such an initiative as the integration, connection and modeling of complex data on Roman road networks in the digital domain was indeed quite innovative in 2005. Had this venture been undertaken manually, i.e. without employing any digital methods or tools, it would undoubtedly have taken longer and it would most probably not been as accurate. Correlating the data so that they become meaningful and usable for my analysis and making more realistic calculations over geographic space, could not have been performed as efficiently relying on analogue methods alone. However, an analogue approach would still have been up to date and re-usable, unlike my 2005 and 2010 end-results. Sustainability of my Roman roads modeling project has proven to be a great challenge, as BetaCMS is not supported any longer, while ArcGIS is not an open source tool. Therefore, one could argue that, what the digital so generously offered my work, it has taken it back rather fiercely.

In this short presentation I will be going through my methodology and results of the Roman Crete land communication modeling and will be raising questions as to their usefulness, curation, re-usability and sustainability as well as the implications to Humanities research, also looking into the prospect of the employment of current innovative digital methods and tools that are available openly. While there is a considerable number of studies regarding digital preservation strategies and planning on an institutional level (e.g. libraries and archives), my intention is to address those very issues from the individual scholar’s perspective.},
howpublished = {Short paper presentation at DH2019 Conference, Utrecht University 9-12 July},
keywords = {},
pubstate = {published},
tppubtype = {presentation}
}