Introduction to long-term preservation

The long-term preservation of DSE is still, as of 2024, an only partially and in some respects unsatisfactorily solved challenge. This chapter can therefore only speak of standard solutions to a limited extent and sometimes serves more as a guide to the direction in which technical developments for long-term preservation are currently moving, which solutions can be expected from these developments and which limitations or planning steps they entail.

In principle, the topic of long-term security can be divided into two aspects, as is done in the following sub-chapters. This distinction, which in some cases does not exclude fluid transitions, is explained here by way of introduction.

1. long-term preservation of the presentation

Long-term storage of the presentation means that the front end remains accessible in the long term. This fulfils a wish of many DSE projects, some of which invest large resources in the design and functionalities of their frontends, even if they are technically standard solutions such as the TEI Publisher. Projects regard the presentation as an integral part of the publication and often describe it in analogy to the printed book (which has so far far far exceeded the longevity of DSE frontends by far).

However, many projects do not have the infrastructure that allows them to maintain dynamic web applications (such as those generated by the TEI Publisher up to version 9) in the long term. By dynamic web applications we mean programmes that generate the desired display each time a website is called up or a search is performed on it with the help of a database. Project durations are usually limited to a few years; hardly any funds are allocated for the project's own maintenance of a dynamic web application beyond these durations. Normally, project funds for long-term preservation are used solely for archiving data records (see below), the funds are therefore tied to external institutions for digital archiving (in Switzerland, for example, to DaSCH).

Research and memory institutions, on the other hand, do not have the long-term preservation of dynamic web applications as a core task, especially as this would entail technical uncertainties and financial risks. To our knowledge, no institution in the German-speaking world currently officially guarantees the long-term maintenance of DSE web applications for DSE projects, even if they have been de facto maintained for longer in some cases.

The solution to this challenge discussed in the subchapter is the generation of static_presentations of DSE, which are currently being (further) developed in various project contexts. They differ from dynamic web applications in that most views are pre-generated in such a way that no further programmes or databases are required in the background. However, solutions are also possible in which certain dynamic functions are retained and, if maintenance is no longer possible, can be switched off without losing the static functions.

2 Archiving the data

At the centre of data archiving is the long-term accessibility of the edition data. This can include a generic presentation of the data. However, the focus is on the long-term accessibility of the data itself, its findability and comprehensibility. Informative documentation and well-formed TEI/XML data are therefore a prerequisite for any form of data archiving. The long-term preservation of presentation interfaces has completely different requirements and is therefore independent of data archiving.

In addition to archived complete data sets of DSE, the archiving of different data types for different forms of use is central. These forms of use generally deviate from the reading and philological or hermeneutic research of the text corpus made possible by the front end; instead, they understand the text as a data basis, e.g. for distant reading or computational linguistic methods, or use it for data visualisations.