Session Summaries by Kenan-Korac

Data, Metadata and Tropy, 25-09-2024

The main topics of the last course given by Sofia Papastamkou were about data and metadata, followed by a brief introduction of tropy. The lesson was divided into three categories. The first category delt with the definitions of data and metadata. What is data? The notion of data changed its meaning over time. After the 18th century the definition of data extended beyond mathematics and theology, the use of which became more generalized. In the 21st century, becoming more and more connected and digitalized, witnessed the emergence of historical data. Historians can produce their own (historical) data which can be manifold and versatile. So in short data are things that we create, collect, store, analyze etc,… Research data are factual records ( images, sounds, maps,…) used as primary sources for scientific research, accepted by the scientific community and seen as necessary for the validation of research findings. Research data can differ between two types of research data: qualitative and quantitative data. The second topic of interest of the course was metadata, which is to put it simply ‘information about information’. Metadata can be useful in order to organize, structure, identify,… data. The explanation of metadata was illustrated by some examples of metadata, like the analysis of image files and Instagram posts. The last part of the course was a short introduction of Tropy, a specialized software which enables you to organize and describe photographs of research material ( photos).

title: Session Summaries by Kenan-Korac abstract: Summary-2 authors:

• Korac-0210945149 date: 2024-10-14 —

Web Archives, 02-10-2024

The class from 02.10.2024 was given by Sofia Papastamkou and it was about web archives. The course was structured in two parts. In the first part the students made groups of three and could choose an assignment. In all there were seven different assignments, but all the assignments focused on a different type of web archive. Afterwards the students were given approximately fifty minutes to answer some questions. The questions were in relation with the web archive an focused on the function, the policies, the methods and limits of those. The latter part of the session was dedicated to a five-minute oral presentation from each group about their findings of their web archive. In my case it was assignment 2- Luxembourg Web Archive. The Luxembourg Web archive is a project of the BNL (Bibliothèque nationale du Luxembourg). One of BNL’s missions is to collect and preserve all publications issued in Luxembourg, as well as foreign publications in relation with Luxembourg. The BNL collects not only printed media, but it also covers digital publications of all kinds. Even Luxembourgish websites. All websites with an ‘’.lu’’ domain, published, hosted or in relation with Luxembourg are being snapshotted 4 times a year and archived. Nevertheless, BNL’s web archive project is relatively new and therefore also limited in some aspects. YouTube videos, podcasts or music videos are not yet archived. To draw the bottom line BNL’s web archive project is a useful and promising tool for everyone interested in research.

title: Session Summaries by Kenan-Korac abstract: Summary-3 authors:

• Korac-0210945149 date: 2024-10-14 —

Online Newspaper Archives& Tool 3 ‘impresso’, 09-10-2024

Course number 4 was held by Marten During, and the course was named ‘Online Newspaper Archives – Tool 3 ‘impresso’’. The course was structured in two parts. The first part introduced us to the world of newspapers, their role in historical research, and the importance of archiving newspaper collections. Besides emphasizing the value of newspapers, we were also introduced to impresso. The second part of the class was the practical application of the theoretical concepts. Returning to the first half of the course, the importance of newspaper collections lies in the fact that newspapers themselves can be utilized in various ways. They can serve as primary sources for research because they offer a broad and diverse range of information about the past. For example, they reflect ideas, past knowledge, and the ways people used to think, speak, and dress. In other words, newspapers mirror past societies. Furthermore, one can observe the evolution of printed media and how writing styles have changed, including how advertisements were once crafted. Newspapers also provide visual support by depicting people, politicians, and so on. For these reasons, we were instructed in and shown how to use the impresso app, which is a newspaper archive that collects Luxembourgish and Swiss newspapers. impresso allows users to utilize archived newspapers in a variety of ways. For example, impresso can help us to segment newspapers and make them more accessible for historical research. Another feature of impresso is that it transcends national borders by storing not only Luxembourgish newspapers but also Swiss ones, and in multiple languages. The second part of the class was dedicated to experimenting with impresso and using its various options, such as ngrams, search, text reuse, and inspect & compare. The goal was to learn how to critically analyze the results generated by impresso.

title: Session Summaries by Kenan-Korac abstract: Summary-4 authors:

• Korac-0210945149 date: 2024-10-22 —

Maps, 16-10-2024

The course on the 16th of October 2024 was about maps and was given by Eliane Schmid. I worked with three other students on the John Snow map. It was the first time I created a story map, and I was surprised that it was not as difficult as I had always imagined. The John Snow map was produced by Dr. John Snow. His map shows the cholera outbreak in London in 1854. He was the first to recognize that the disease did not spread through contaminated air but through contaminated water. Moreover, his map is a good introduction to GIS, as it allows the cartographer to experiment with different layers. We used only five layers, which helped us a lot because we were only dealing with a manageable number of layers. This allowed us to fully understand the function of each one. Additionally, I learned how to create a story map myself and how to apply different layers to highlight various aspects. For example, the water pump layer displayed all the water pumps in the area, which helped us connect the dots between the death toll addresses and the pumps. While making the story map, I also became convinced that using maps in general helps narrate a story more clearly and understandably. The interplay between text and maps engages the reader, making the experience more playful and interesting. Overall, it was a good first introduction to creating story maps.

title: Session Summaries by Kenan-Korac abstract: Summary-5 authors:

• Korac-0210945149 date: 2024-10-23 —

Networks & Tool 5 ‘Palladio’/’Vistorian’, 23.10.2024

Course number 6 covered networks and tools like ‘Palladio’ and ‘Vistorian’, which are specialized for creating, or rather visualizing, networks. The advantage of network visualization is that it helps the creator see their information more clearly, which can be very helpful, especially because they may identify networks or connections they hadn’t previously noticed, as well as other details like the density and different types of networks. In other words, network visualization can reveal overlooked aspects or patterns that might not be immediately apparent. Another significant benefit is that networks can be applied to any topic, allowing any subject to be represented by a graph. Network visualization also offers the ability to query what is not shown, helping to understand why certain elements are missing from the display. In network analysis, theory and method are not separate, which can be challenging, especially for students encountering the material for the first time. Fortunately, our professor explained the theory, terminology, and concepts used in network analysis, such as ego network, broker, unipartite network, and bipartite network, through practical demonstration at the start of the course. He illustrated these concepts using the example of a wedding, starting with the creation of a guest list, then moving to seating arrangements, and finally the table conversations. Through this example, he effectively conveyed the theory and terminology behind network analysis in a simple and understandable way.

title: Session Summaries by Kenan-Korac abstract: Summary 6 authors:

• Korac-0210945149 date: 2024-10-30 —

EP Archives, Presentation, Queations, 30/10/2024

The Archive Dashboard Unit was launched in 2021 with the sole purpose of “democratising knowledge of the history of the European Parliament”. The archive consists of over 2 million digitised documents from the Common Assembly of the European Coal and Steel Community, the Ad Hoc Assembly and the European Parliament. There are four different dashboards. The first is the overview dashboard, which provides a general view of all digitised documents from 1952 to 1994. There are currently 2,142,682 documents available in the archive. The overview dashboard also shows the language of each document, indicating whether it is in English, French, German or other languages. Another feature of this dashboard is that it shows the different types of documents, allowing researchers to filter by type, such as letters, questions, applications or oral tapes. A key feature of the overview dashboard is that it shows the year in which each document was created, allowing users to see how many documents were created in each year. It also provides insight into the evolution of document growth over time.

The second dashboard is the Content Analysis Dashboard, which, as the name suggests, allows for more in-depth analysis of the documents. This dashboard offers several tools that make the documents more accessible and help researchers to explore the history of the European Parliament more effectively. The first tool, “Top Words”, provides insights into the themes of European motions and questions. Its interactive feature allows researchers to click on a word of interest and view all documents containing that word, while the second tool, ‘Dominant Topic’, offers a closer look at the development of topics over the years. In addition, the Content Analysis dashboard allows researchers to perform even deeper analysis by providing similar documents on a given topic. The third dashboard is the Archive Requests Dashboard, which uses an interactive world map to show the countries that have made requests to the Archives Unit. In addition to countries, it also shows which organizations, institutes, research laboratories and others have made requests. The dashboard shows the number of requests and the year they were made. The fourth dashboard offers a summariser that allows the user to create a summarised text from Word or PDF files in any language. It is also possible to select the length of the summarised text. This is great for getting an insight and overview of the text. In conclusion, the Archives Dashboard Unit is a very useful tool for researchers and individuals interested in the history of the European Parliament. Its data visualisation makes information easily accessible, while the specific functions of each dashboard allow for deeper analysis of documents. Features such as keyword search make research more efficient and valuable for the researcher. However, it does have some limitations, as it only provides an insight into the Parliament’s archives; a big plus would be if it also provided access to the archives of other European Union institutions. And, of course, any model can have hallucinations. The discussion after the presentation centred on the liability of the documents retrieved by the EP Archives system. Any basic model can have hallucinations and therefore a small error rate. Nevertheless, the systems provide all the most interesting documents in relation to the query. Although the system has a summariser that summarises the main components of the documents, the documents themselves are also provided. So a researcher would never fully trust the summary, but would always access the documents themselves and critically analyse them and then draw his or her own conclusions. Other questions related to the possibility that the EP archive model could serve as a model for other institutions, such as the National Library, to improve and facilitate access to the archives.

Questions prepared before the session.

1) How reliable is the information provided by the ‘Ask the EP Archives’ tool? 2) One of the tasks of the Archives Unit is to ‘democratize’ the EP archives and make them accessible to citizens. Do you think that your system is simple enough for ordinary citizens without any IT training to access the documents themselves? 3) Are you planning to extend the Archives Unit to the archives of other European Union institutions? 4) Is the ‘Ask the EP Archives’ tool modelled on ChatGPT? 5) The Archive Unit provides free access to the EP archives. Are there any restrictions, i.e. if you want to access a document, can you access all the documents online or do you have to go physically to the EP archives?

title: Session Summaries by Kenan-Korac abstract: Summary 7 authors:

• Korac-0210945149 date: 2024-11-06 —

DH:Theory,Criticism;Transparency;Reproducibility/Documentation, 06.11.2024

The course began in a rather unusual way, as it was the day of the US presidential election. And we witnessed the second election of Donald Trump as the new President of the United States. During started the course by discussing our views, the students’ views, on the election of Donald Trump. After twenty-five minutes of discussion, During made a smooth transition from Donald Trump and fake news to DH theory. The question we were asked was ‘how do you know what is true’? As an example, During mentioned David Irving, author of several controversial books, e.g. The Destruction of Dresden. In his book he exaggerates the destruction of Dresden, and his thesis is based on forged documents. He later even denied the Holocaust. Irving has been heavily criticized by several historians such as Deborah Lipstadt. Nevertheless, the main question remained. How to prove that Irving had deliberately and knowingly falsified archival documents. After years of examining his works, speeches, notes and so on, investigators were finally able to prove his evil intent to deliberately manipulate documents. This led to the next question. How do you do this for data-driven research? How do you document your data so that it is considered valid and can be reused and reworked? This question can be answered by meticulously documenting the steps we take during our research. We need to explain why we include certain aspects in our research and not others. Put simply, we need to explain our selection and the choices we have made. To illustrate this theory, we used the networks we had created the week before. We ended the session with the F.A.I.R. (Fidable, Accessible, Interoperable, Reusable) principles that every historian should follow.