Projects Hackathon Coding Gender 2019

In the summer of 2019 the hackathon „Coding Gender – Women in Cultural Data“ took place in the Berlin State Library. During this three-day event, interdisciplinary teams of computer scientists, designers and humanities scholars developed innovative projects using digitised historical documents from the Berlin State Library and addressing questions such as the visibility of women in cultural data, the construction and representation of gender roles and the relation between today’s socio-political debates and historical gender stereotypes.

Short descriptions of the projects realised during the event and awarded by the jury can be found below. The topic data sets created from the Digitised Collections of the Berlin State Library that were used during the hackathon are accessible here.



Benimm-Dich-Bot (Behave-bot)

The historical Twitter bot “Benimm Dich, Du Frauenzimmer!” (“Behave, woman!”) has advice for every life situation!

The Benimm-Dich-Bot gives advice for small and big problems in the life of modern women*: But beware, it should be used with caution. Its role models date back to the 19th century and its credo is “Behave, woman! The advice – or rather, rules of conduct – were found in the digitised collections of the Berlin State Library from 1830 to 1930 as part of the Hackathon Coding Gender. The bot allows a playful examination of the problematic gender roles of the time.

For this purpose, corresponding text passages from the digitised collections were identified, transcribed and assigned to at least one of the categories #Career, #Household, #Family, #Love, #Sex, #CoupleGoals, #EverydayLife, #FemaleTroubles, #Beauty. If you ask the Benimm-Dich-Bot on Twitter for advice using one of these hashtags, a suitable text passage will be selected from the collected texts and posted with a link to the respective digitised version as an answer.

The underlying technology is based on the Python library Tweepy and allows you to quickly implement your own ideas according to this example. As the extracted text passages are sometimes too long for the tweets (or the SBB and Twitter short URLs), the link to the digitised material is currently sometimes cut off. The complete texts and URLs can also be found in the JSON file used internally by Benimm-Dich-Bot.

Try it yourself under:
Advice given so far can be found in the replies.


Lou Andreas-Salomé – Interactive Storytelling

Extraordinary find at the Hackathon Coding Gender: the mobile phone of Lou Andreas-Salomé! She was smart, popular and incredibly well-connected, the psychoanalyst and writer Lou Andreas-Salomé; a true influencer of her time. But in the black and white pictures we can see of her today, she seems rather aloof, withdrawn and serious. Andreas-Salomé wrote numerous narrative texts, studied with Sigmund Freud and then did her own research in the field of psychoanalysis. But today hardly anyone knows her for this. Instead, she is usually remembered as a kind of muse of other intellectuals, since as a young woman she had close relationships with Paul Rée and Friedrich Nietzsche, and later with Rilke.

Storytelling: From Muse to Influencer

Lou Andreas-Salomé, c. 1897. With blind stamp of the Elvira studio, Munich

The project “Lous Handy”, which was launched at the very first culture hackathon Coding Gender of the Berlin State Library, aims to cast a different perspective on the woman who has shaped her life so self-determined and free. Lou’s life is just one example of a woman’s life, which can be prepared with the help of alternative storytelling like the one developed here for Lou. Thus, the prototypical project result is only a first step into the data set “Individual Lives – Documents of Individuals’ Lives”, which was made available by the the Berlin State Library during the hackathon.

Letter? WhatsApp!

To be able to tell her story in a new and different way, the team read her letters to Gerhard Hauptmann and his wife Margarete, deciphered her handwriting and turned them into digital texts. This includes the letters that Friedrich Nietzsche wrote to her as well as digital versions of her stories, novels and factual texts. Also the biographical texts about her according to the people with whom she cultivated friendships, from whom she was intellectually inspired, whom she loved and who loved her were searched extensively. The prototypical app, which was then developed and the story it tells, want to lead the users into her work, starting from her person. For this purpose, the bundled data was further analyzed. A spectrum of methods from the Digital Humanities was used, including network analyses with Gephi, Topics created with the DARIAH-Topics-Explorer and stylistic investigations with the R-Package Stylo.

The result in a screencast

The story of the project presentation hides all the hustle and bustle of the 1.5 day hackathon as best it could. After all, it was not the participants who were to be the focus of attention, but Lou Andreas-Salomé. And so here the team itself tells you about how they found Lou’s mobile phone by chance. See for yourself:

What’s going to happen to Lou’s phone now? The project focuses mainly on alternative storytelling. The reconnection of the reality of intellectual women’s lives to their work and also the portrayal of women as independent intellectual personalities remain an urgent concern even after the Coding Gender Hackathon. “Lous Handy” will probably be integrated as a prototype into one or more university teaching projects, in which students of literary studies themselves can tell the story of women poets and thinkers in a different way.

Frawenzimmer. The new magazine!

The project “Frawenzimmer” was created as part of the Hackathon Coding Gender. It takes up the current genre of women’s and gossip magazines and combines it with historical sources from the digitised collections of the Berlin State Library. A large part of the topics and discourses currently under discussion have a long tradition! The tone of voice is also crucial, because what often appears sober and emphatically factual in historical sources serves gender orders and power structures of the time. Thus, on a supposedly scientific basis, attempts are made to legitimise inequalities between the sexes.

Old discourses in current women’s magazines

One of the most important observations was that they are structurally not very different from current women’s and gossip magazines, and that these are still being updated today. Topics included: toys marketed in a gendered way, focus on clothing in sports performed by women, topos of the greater sex drive in men and related acts of sexualised violence, the myth of guilt in cancer, tips for sex, different expectations of mothers and fathers, the role of women as carers, gender pay gap, etc.

Title page of “Frawenzimmer” – click on the red, pulsating dots to call up the corresponding digitised images

By imitating the lurid tone and linguistic style of gossip magazines, the aim is to make the historical texts accessible to a broader target group and thus encourage to discuss them. This also reveals a further correspondence between the selected historical texts and the format of today’s women’s magazines: as texts written about or for women, they are often of a prescriptive nature: they are instructive, give (well-intentioned) advice, and explain what women should do and what they should not do. The project was intended to make these continuities visible and also to facilitate access to the historical texts by means of so-called teasers – as they are also used on the covers of gossip magazines.

The articles and illustrations in the “Frawenzimmer” are themselves excerpts from the historical sources, only the headlines and layout have been newly created and are based on the genre of women’s and gossip magazines. Each article contains the information with a direct link to the digitised objects of the Berlin State Library.

Beruf-O-Mat. Find your former self

Based on the data sets provided for the Hackathon Coding Gender, the Beruf-O-Mat is an exciting quiz that uses six questions to find out which CV would have led to which (female) profession in the past and how well (or badly) women in the different eras were able to make ends meet with it.


1. Area: Where were you born? (City / Country)

2. Time: When did you grow up? (1870s / 1890s / 1908 to 1914 / 1914 to 1918)

3. Social background: In which social class were you born? ( Upper class / Upper middle class / Lower middle class / Lower class )

4. Education: What school did you go to? (elementary school / private lessons / citizen school)

5. Secondary school education / study: Where did you go to secondary school? (secondary school for girls / private school for daughters or boarding school for girls / teacher training seminar / school for arts and crafts / grammar school for girls / study as an auditor without a degree / study with a degree)

6. Marriage: Are you going to marry? (Yes / No)

The combinations of answers then lead to professions such as factory worker, maid, nurse or even writer, journalist and doctor.



More detailed information can be found here: Berufomat

A first implementation of this quiz can be found under the following link (Attention: Link only available for some time!)

Weibsbilder. World Wide Women.

How can role models be represented? What relationships exist between specific role attributions, the activities performed and the source genres? Are there congruences between specific roles and source genres? These questions were answered by the team “Weibsbilder. World Wide Women” in the Hackathon Coding Gender. Based on the data sets Picturing Gender, Dolls Kitchens and Aeroplane Kites and Women in Wartime, these questions were investigated using the graph database neo4j. The available data sets were used to show how women were portrayed in different media at different times and what role attributions were made to them. At the same time, the use of neo4j in conjunction with Wikidata was intended to demonstrate the possibilities of indexing and mediation in order to give more attention and visibility to women and role images in historical sources – in line with the theme of the hackathon “CodingGender”.

  1. As a first step, the available data had to be extracted manually from the data sets, from which the property keys “Person”, “Occupation”, “Description” and “Role Assignment” were created. For a possible evaluation of the source genre, also the property keys “Medium”, “Publication”, “Title”, “Subtitle”, “Author” and “Identifier” were created.
  2. In the second step, the relationships between the nodes were marked. The respective source was described with the identifier of the Berlin State Library’s Digitised Collections in order to be clearly identifiable and verifiable. The descriptions of the women and their role attributions had to be lemmatised before being entered into the database.
  3. In parallel, the sources were entered as data records (also manually) in Wikidata and Wikimedia with their corresponding statements (inventory number, GND-ID etc.). The identifiers of the digital collections of the Berlin State Library were also used for the individual sources. The respective unique identifier from Wikidata was then assigned to the respective source in neo4j.

With just a few data sets, the Hackathon project “Weibsbilder – Word Wide Women” wants to show the possibilities of using graph databases in combination with Wikidata: On the one hand the use as a research environment and on the other hand the increase of visibility of women in the World Wide Web. If the project is continued, the development of a user interface would be desirable, both to facilitate the communication of the research results and also to enable the addition of further data/data sets, for example in the context of a Citizen Science project.


Digital library (DL) support for different information seeking strategies (ISS) has not evolved as quickly as the amount of content or the quality of presentation. However, several studies argue for the support of explorative ISS alongside the directed query-response paradigm. Hence, this project presents a primarily explorative research system prototype for metadata harvesting allowing multimodal access to DL content for researchers during the research idea development phase, i.e., while the information need (IN) is vague. To address evolving INs, the prototype also allows ISS transitions, e.g., to OPACs, if accuracy is needed.

As its second contribution, the project presents a curated data set for digital humanities researchers that is automatically enriched with metadata derived by different algorithms including content-based image features. The automatic enrichment of bibliographic metadata is needed to support the exploration of large metadata corpora as traditional metadata does not always address vague INs.

This demo clearly shows that use-case-specific metadata facilitates the interaction with large metadata corpora.

When clicking the following link, make sure you run Chrome or Firefox as we deny the existence of other browsers. You should also own one of the fonts ‘Helvetica Neue’ or ‘Helvetica Light’, use a reasonably fast computer as the next page makes heavy use of JavaScript, and have a fast internet connection to quickly load a lot of images. You should also remember to click the ? if you have no idea what to do.

Enter Demo

Double-clicking a document lets you inspect a cluster. Clicking a document once makes it the current document.

Back to overview
ESC Remove comparison edges
b Build metadata package
c Compare current document to others
d Display details of current document
m Display map of displayed documents
p Pause animation
s Start animation


Source code

Developed by

  • David Zellhöfer



ShiCo – Exploring Shifting Concepts Through Time in Digitized Newspapers.

ShiCo is a tool for visualizing semantic shifts in certain concepts over an extended period of time. “Concept” here means a set of terms which are semantically related to a central subject term. ShiCo uses a set of semantic models (word2vec) spanning a number of years to explore how concepts change over time – words related to a given concept at time t=0 may differ from the words related to the same concept at time t=n.


Source code


Developed by

  • Netherlands eScience Center
  • Institute for Natural Language Processing University of Stuttgart

Berliner Schlagzeilen (aka “Berlin Headlines”)


100 years ago, the Berliner Volks-Zeitung was the daily newspaper for the metropolis of Berlin. With a mix of serious high-brow content, light entertainment and sensational pieces it became the most successful tabloid newspaper of its time. The Twitter bot “Berliner Schlagzeilen” tweets the headlines and a picture of the title page from 100 years ago every day, on the day. In this way we connect the past with the present and make history accessible to a broad audience.

This application was developed during the event Coding Da Vinci Berlin 2017. The aim of the application is to make digitized historical newspaper articles easily accessible for users. In cooperation with the Berlin State Library, digitized newspaper articles were integrated into a Twitter bot.


Source code

Developed by

  • Falko Krause
  • Erik Koenen
  • Olivier Wagener



Altpapier (aka “used paper”)


Read news from 100 years ago. With this app you can discover random newspaper articles from the days before yesterday. Find your favourites, such as quirky anecdotes, old personal ads or local phenomena.

The Altpapier (aka “used paper”) app brings the most exciting and unusual newspaper reports from the beginning of the 20th century to your smartphone. The app accesses digitized historical newspapers from Berlin State Library and makes them available via a modern news app. The app offers various feeds; individual articles are displayed with the date, the name of the newspaper, the headline and the actual news item.

This app was created during the event Coding Da Vinci Berlin 2017. The aim of the application is to make digitized historical newspaper articles easily accessible for users. In cooperation with Berlin State Library, digitized newspaper articles were integrated into an Android app.


Altpapier can be downloaded from the Appstore.

This app was created using Android Studio. A version for iOS is under development.

Source code

Developed by

  • Maria Becker
  • Erik Koenen
  • Kathrin Konkol
  • Maike Orlikowski