Interview Oliver Schmid
Dr. Oliver Schmid on Translating between Computer Scientists and Researchers in the Humanities and the Nature of Beta Versions
What do you do at the Centre for Electronic Text Processing and Publication in the Humanities at the University of Trier?
Schmid: As an intermediary between computer scientists and scholars in the humanities, one of my main tasks is translation. I provide users with information about changes from the programming side that could be relevant for them in a language that they understand. I also ask the users about their needs and pass on any modifications they request to the developers
As an economic informatics specialist, was it initially difficult for you to adapt to the language of the humanities?
Schmid: There were few comprehension problems concerning the technical problems that are our primary focus of communication. This is probably because here at the Centre I mainly deal with humanities researchers who have worked with digital methods for a long time and thus to some extent are familiar with computers, even if it is not their native territory. Nevertheless, linguistic misunderstandings can occur. You have to be careful with terms such as “editor”, because for humanities researchers it primarily means a person who works in publishing, while for computer scientists it refers to software for data processing. Recently we developed a way to combine digital objects from different projects, and we wanted to call this combination a collection. In our collaboration with librarians, however, this caused some communication problems because in the library context, this term is used for a compilation of works by a single author.
What is it like when you interpret?
Schmid: Mostly I translate technical terms into normal everyday language and make technical processes comprehensible through metaphors, for example. The projects and objects in TextGrid can be explained by comparing them to directories and files in an operating system, or I might try to translate if someone has difficulty with terminology in the “Help” function that is actually intended to be helpful, such as the term “cross-domain single sign-on.” It is less than helpful if the user does not understand what is being discussed. This term means something very practical: Normally, users have different passwords on different servers and sometimes they have problems deciding which one is correct. For this reason, with “Single Sign-on” users have the ability to tell all servers from participating institutions that no matter where users log on, their password is always the same – just like a master key. It is also available for the TextGridLab. With single sign-on, users do not have to worry about which server they are accessing, and with their passwords they can access all of the services and search for data as well as use certain functions of the software. One task that we have yet to tackle is the formulation of the “Help” function in such a way that even someone who is not a computer scientist will be able to understand it completely.
Do you also work with the tutorials, the instructions for the programs?
Schmid: Yes. We have now begun to develop a dictionary in order to systematically use the same words and to avoid alternating between synonyms such as “folder” and “directory” so that users do not have to think about whether one synonym has a slightly different meaning.
Do users normally contact you with a request, or is it more common that the developers show you new techniques so that you can help them discover how they can be used?
Schmid: Both can happen. We also meet with both sides to discuss specific areas of application. One example is our Text-Image-Link Editor: In our meetings we discuss which capabilities the software should have. The philologists describe their work to us and ask concrete questions, such as whether the document must always be aligned in one direction relative to the screen. They might ask if it could also be displayed at an angle, or if there could be marginal notes, such as are found in many manuscripts. There are also writers who reach the bottom of a page, rotate the paper, and continue to write on the margins. Experts in their field, such as philologists, are the only ones with an overview of the wide range of such possibilities, which is why all approaches that are implemented need to be thoroughly discussed with the computer scientists. We were recently asked about the options for a specific project dealing with the Text-Image-Link Editor. The project members are interested in simultaneously displaying multiple images, rather than one picture at a time, so that users could flip through several scanned images on separate pages without having to close and open each document separately. This was a request we had never thought about before. Now we can think about what is technically feasible, and how we could implement it, so that the user will have an intuitive interface.
In the intuitive interface for the beta version there is probably still work to be done...
Schmid: We have to hope that it is clear to the users that the TextGridLab is in the development process and therefore the software is not yet finished. There are still occasional errors and the menu interface is not yet perfect. We are currently planning to adapt the user interface so that the various tools are more usable. The menu structure also contains items that are relevant to the programmer but not to the user. These things appear awkward and unripe in a beta version. This software is in a continuous stage of development and improvement, and for this improvement we need regular user feedback.
How often does a new version come out?
Schmid: Every two or three months. The most current version will be announced on our homepage, along with a link.
Interview by Esther Lauer.