This was ER 5: brief overview of my DiXiT activities

This blog post summarises some key aspects of my DiXiT research and looks back at events and activities I organised or participated during the 12-month of my fellowship.

The title of my DiXiT fellowship was ‘Canonical reference & sustainability of digital editions’ and while I initially proposed to address issues related to persistent addressing of interactive content in digital editions, I gradually invested more and more the topic of canonical referencing and persistent granular addressing in digital editions too. The reason is that I am working on a documentary edition of St Patrick’s epistles and questions of canonical referencing and granular addressing are central to this edition and it provided me with a case study for my research and a playground for experimentation.

My host institution

The host institution for my DiXiT fellowship was the Centre for Information Modelling in the Humanities – Austrian Centre for Digital Humanities (ZIM-ACDH) at the University of Graz. The ZIM-ACDH is a research centre that exists since 2003 with a strong focus on Digital Scholarly Editing and digital preservation. The ZIM – ACDH was the perfect place for my research, as the GAMS, a Fedora-based OAIS-conform asset management system, is developed and maintained here. In the GAMS a number of digital editions have been and are being developed such as the Urfehdebücher der Stadt Basel – digitale Edition and my DiXiT colleague Frederike develops her edition Stefan George Digital. These are editions based on very diverse sources and with different goals and editorial approaches and there are several other editions in the GAMS that are again very different. The study of these different editions and editorial approaches gave me important input and insights for my research. Furthermore, the ZIM uses Handles for persistent addressing. I could profit of the already existing in-house expertise. My colleagues were genuinely interested in what I was working on and I had many fruitful discussions about my research topic. Special thanks to my DiXiT supervisor Georg Vogeler and Johannes Stigler, the head of the ZIM-ACDH, for their continuous input and support.

Stable references and digital editions

Over the last year I looked into different aspect of scholarly referencing in the humanities with a particular focus on the referencing of resources in digital editions. Persistent addressing of digital resources in general and digital editions in particular is an highly important topic for everyone working in the humanities these days. Referencing and citing is at the core of humanities scholarship and editions are important as they are supposed to provide stable reference texts for unambiguous and persistent citation. However, in digital editing the reality is different, Often it is difficult to know what part of a resource provided by a digital edition can be and how a resource should be referenced so that another researcher in five, ten or 50 years can identify and retrieve the resource (or parts of it) correctly.

Looking at referencing in digital editions there are a great number of topics that have to be investigated (and every time I explored one of these topics, new topics came up). A central area of investigation was the nature of references and citations in the humanities (and digital humanities) and how they are used in print and digital contexts. Important is to distinguish between the two core functions of a reference: identification and addressing. A reference in a scholarly work can be seen as an identifier that allows to identify a publication of some sort. The actual addressing happens in a library catalogue where a shelfmark provides access to the publication itself. The FRBR model provides a very useful framework to clarify these difference. Identification of a resource happen usually on the expression or manifestation level, while access happens on the item level. Special cases of referencing in the humanities are canonical reference systems which are in place for many historical texts and the reference of manuscripts and other unique objects.

What changes when moving to the digital? First of all, the access of digital resources is a different one. Instead of real-world access points such as library or archive, digital resources are globally available via the Internet. And indeed in the last two decades more and more humanities data, articles, images of manuscripts, etc. have become available online. While this instant access of research material is certainly a huge benefit for researchers, the downside is that digital resources are still quite fragile. For instance, they can be removed or replaced very easily or become unavailable for some other reasons what results in the so-called ‘link-rot’ problem. Another problem is that often digital resources are not fixed publications, but change over time. Sometimes changes are only corrections of a typo, but sometimes it can be major rewrites of a digital publication. In both cases a new version is produced. However, when a new version is made available, often previous versions are replaced by this new version and any reference to the previous version (e.g. via a hyperlink) points now to the latest version which might be very different.

An obvious way to start research into stable referencing is to explore and compare the various PID systems (such as Handle, URN, PURL) use in digital editions. I went through the catalogue of digital editions published by Patrick Sahle (http://www.digitale-edition.de/) to investigate how digital editions recommend to cite objects in the edition and if stable references (e.g. permalinks) are provided. Up until 2009 the provision of permalinks or reference guidelines was the exception, however in the last decade more and more digital editions provide PIDs for some of their resources. However, it is not always clear what is represented by a PID. Interfaces of digital editions can be fairly complex and might show a transcription, an image of a manuscript, metadata and maybe even graph visualisations side-by-side on the same page. These are all features that could be cited and, therefore, before implementing a technical solution, it is necessary to identify what resources of the edition need stable PIDs and on what level of granularity should they be assigned. For instance, is it necessary that every word in a text can be addressed with a PID or is it enough to assign PIDs to paragraphs or chapters only?

During my research into granular addressing I investigated the Canonical Text Service (CTS) and the CITE Architecture. My interest in this approach was triggered by the fact that Perseus invests currently a lot of resources to make ancient texts citable via CTS URNs (http://cts.perseids.org/). While i find the CTS solution very interesting, it suffers the same fundamental issue of all URNs, namely URN resolvers are not interconnected and in order to resolve a URN the address of the correct resolver has to be known. An additional (and even more fundamental problem) is that the namespace CTS for canonical text services was never registered with IANA. Only this registration would, at least to some degree, guarantee global uniqueness, one of the fundamentals URN idea is based on.

Another question that I investigated was the role of the interface and interactive content (such as interactive charts and graphs) in digital editions and if it is

necessary to provide stable references for these features. Examples of such complex content can be found in the ‘Das altägyptische Totenbuch’ or the ‘Jahresrechnungen der Stadt Basel 1535 – 1610: digital’ (see the images below). How can stable references be provided such complex interfaces and visualisations, and should such features be made citable? It could be argued that only the underlying data is relevant. But is the choice of a particular visualisation or the presentation of text and images in the interface not important to be preserved?

In regard to providing references to particular interface settings experimental work is being carried out in context of the development of the new EVT 2.0 tool by Roberto Rosselli Del Turco and his team. A new bookmark functionality generates a ‘direct reference to the current view of the web application, considering view mode, current document, page and edition level, eventual collated witnesses and selected apparatus entry’ (link). Roberto Rosselli Del Turco and Chiara Di Pietro presented about the new EVT at the DiXiT conference Digital Scholarly Editions as Interfaces (to the conference report).

Conferences, workshops and other activities

One of the highlights of my year as a DiXiT Fellow was the opportunity to participate in DiXiT events and major DH conferences. I could attend the DH 2016 conference in Cracow. It was my third major DH conference and a great opportunity for networking and to see what other people in the DH community are working on. Several papers and projects that were presented at the conference where directly relevant for my own research (for instance, The Canonical Text Services). It was also my first opportunity to meet many of my DiXiT colleagues who presented at the conference. I presented papers at the DocEng 2016 conference and the TEI conference in Vienna and at the DiXiT 3 Convention in Antwerp (to the conference report). At DocEng I participated with a paper in a workshop (DocChange) organised by my DiXiT colleague Gioele Barabucci and at the TEI conference I presented papers about the Versioning Machine (with Richard Breen) and the Letters of 1916 (together with my DiXiT colleague Richard Hadden) (link). I also had the chance to present about my ongoing research at informal events such as the LUNCHTIME LECTURES at the ZIM – ACDH or at the CCeH in Cologne.

Besides research and attending of conferences I was also co-organising two conferences: The DiXiT conference Digital Scholarly Editions as Interfaces conference in Graz in September and the Versioning Cultural Objects conference in Maynooth in December 2016. This was a great learning experience and currently I am working as co-editor to produce edited volumes with contributions from these conferences. These books will be published with the Institut für Dokumentologie und Editorik (IDE) .

Secondment in Cologne

In August 2016 I made a secondment at the Cologne Center for eHumanities (CCeH). This was great! I got insights into the projects that are being developed at the CCeH and their strategies for persistent addressing, had many fruitful discussions with colleagues there and I gave a paper about my DiXiT research. In particular, many thanks to Patrick Sahle and Franz Fischer for hosting me in Cologne and for taking the time to discuss my research.

My secondment in Cologne coincided with a meeting of members of the Union of German Academies of Sciences and Humanities at Cologne who discussed issues of persistent addressing of digital resources developed by the German Academies. The meeting was organised by Patrick Sahle and for two days I had the chance to learn about Academy projects, the current situation and plans for future steps to ensure their digital content stays available. I was also invited to a second meeting in Mainz where discussions continued.

What’s next?

Right after my DiXiT fellowship I got an employment at the Kompetenznetzwerk Digitale Edition/Competency Network Digital Edition (KONDE) at the ZIM – ACDH. In the context of KONDE I will have the opportunity to continue my research in the area of digital editing. Additionally, I am finalising a publication which will incorporate the research that I carried out during the last 12 month as a DiXiT fellow. I will also pursue the publication of the two books resulting from the DiXiT conference in Graz and the conference in Maynooth which I co-edit with several of my colleagues.

 

 

---

---