Heinz Kramski, Deutsches Literaturarchiv Marbach, ORCID 0000-0001-5216-278X, kramski@dla-marbach.de, 20. Mai 2026
Colleagues from the Manuscripts Department of the German Literature Archive Marbach (DLA) regularly visit donors/former owners of literary estates and collections on site prior to acquisitions in order to inspect, evaluate, and select the archival material that may be transferred.
Increasingly, they encounter born-digital materials during these visits, such as actively used or retired computers from various manufacturers and all kinds of storage media. However, hardware and born-digitals are difficult to assess at first glance: technical metadata, details of their use, and their contents are not immediately apparent, but instead require tools, user knowledge, and in some cases technical expertise typically acquired through IT-oriented training.
At the same time, such visits offer the opportunity to clarify important details about computer usage and, where applicable, login credentials and access data directly with the record creators.
To support a more systematic collection of data on born-digital materials in acquisitions, DLA's Digital Archive Unit has developed a digital questionnaire. It can serve both as an input aid and as a checklist, helping to structure and facilitate the transfer, future processing, and cataloguing of unique digital materials.
In addition to instructions for use, the questionnaire consists of four fixed sections (pages), which can be completed with varying levels of detail:
- General Information
- Type and Extent
- Content
- Final Remarks
The section “Type and Extent” is especially important: the components of the acquisition listed there (and roughly quantified) determine up to six additional dynamic pages in which specific categories can be described in greater detail:
- Devices
- External and Removable Storage
- E-mail Accounts
- Data Sets in the Cloud
- Other Data Sets
- Other
The “Devices” page, for example, currently allows the description of up to 10 devices, such as desktop computers, laptops, tablets, and so forth. Different levels of detail are possible: each device can either be summarized in a short text describing its main characteristics (type of device, type of use, status within the acquisition, general technical specifications, etc.), or alternatively – if sufficient time is available – documented using structured details.
The other dynamic pages are mostly organized in a similar way.
The questionnaire is available in both German and English. In the English version, some technical terminology may still require improvement.
The DLA questionnaire consists of a single JSON file defining a JavaScript application. It is based on the SurveyJS Form Library (https://surveyjs.io/form-library) by the company Devsoft Baltic OÜ (https://surveyjs.io/contact-us). The Form Library itself is open source and licensed under MIT (https://surveyjs.io/form-library). Only additional functions such as the Form Creator or the PDF Generator require paid licenses. Generated questionnaires may be used and distributed without restriction (https://surveyjs.io/licensing).
The questionnaire JSON file is made publicly available by the DLA on GitHub under a CC BY-SA 4.0 license.
It can either be edited directly with a text editor (“the hard way...”) or more conveniently using the Form Creator, either as a local installation or via the provider’s online demo (https://surveyjs.io/survey-creator/examples/free-nps-survey-template/reactjs).
With around 10,000 lines, the DLA questionnaire appears to be one of the more extensive SurveyJS applications. For larger structural modifications – for example, when entire device sections had to be duplicated multiple times – editing in a text editor proved easier than working with the interactive Form Creator. To nevertheless maintain consistent IDs for the hierarchical sections and questions, and especially for cross-references, the DLA developed a small AWK script SurveyJSTidy.awk that systematically consolidates all IDs.
It is possible to render a questionnaire JSON file locally using an HTML file on a laptop. At minimum, this setup will be demonstrated interactively in September (it is already working now, see Demo).
The demo will use the iPRES Bake-Off Pantry (https://drive.google.com/drive/folders/1_BFjNw95HhH45VO-Y2gmJTbd6kfYAtuY) as an (artificial) example for the Data Sets in the Cloud section.
The online demo of the Form Creator also allows arbitrary questionnaire JSON files to be imported, rendered, and modified. This provides another possibility for interactively testing our questionnaire.
The DLA is currently working on creating a full-featured web application with a database backend and user management. However, for reuse of the questionnaire elsewhere, this environment is secondary and of limited interest because it is too specific to local requirements. Devsoft Baltic offers various examples for backend integration.
The questionnaire is still being tested at the DLA and has not yet been widely deployed. It is already becoming apparent that the level of detail that would ideally be desirable is often at odds with the limited time available during on-site visits, where extensive analogue holdings usually also need to be inspected. All the more, we are interested in feedback from the community.