Reading time: less than 5 minutes.
–> Lire la version française de cet article 
As blog subscribers know, I have been working for several months as part of the Manioc.org team to do the mediation. In particular, I have written several blog articles in French to highlight the collections of this digital library specializing in the Caribbean, Amazonia and the Guyana Plateau. But that wasn’t my only task, far from it! The library also had an ambitious project that was very important to me: a database « Esclavage en Martinique ». For almost a year, I accompanied this exciting project. So for once, I’m talking to you about the background of the job!
The origin of the project
For many years, Mr. de Reynal compiled an incredible amount of data that he found in some 3200 notarial deeds; in his desire to share his work, he wondered how best to do so and contacted Manioc Library, which already hosted a database « Esclaves et affranchis de Guyane » [Slaves and freed slaves of Guyana]. It was proposed to him to create an open access web interface to make the data accessible to as many people as possible. Thus, a sub-domain of the Manioc.org website was created: you can now search in the « Esclavage en Martinique » database indexing more than 3200 notarial acts of the 18th century and listing more than 14700 enslaved people in colonial society in Martinique.
How did we move from collecting rich and complex data to building the website? The steps are numerous and essential to a quality result.
Expertise of the archives and data collected
Before putting data online, it is necessary to understand where they come from and how they were collected. It is therefore necessary to go through an expertise phase where the reliability of the sources used and the data collected are checked.
The expertise of the sources is really a moment I like, because if it is a question of assessing the reliability and value of the content, it is also an opportunity to explore documents that are generally unknown. This is the moment when you discover the history of a documentary collection. And it is not every day that we have in front of us piles and piles of old papers that someone blackened with ink more than 200 years ago! See the signatures, meet familiar names, immerse yourself in everyday life through what was produced as part of people’s activities at the time… In short, it is the moment when we are in contact with the tangible traces of history and when we immediately imagine all that we could learn from it!
After the work on the sources, it is time to start working on the database. I’m familiar with databases! Not only do I use it, but I also structure it. However, if you are not the author of a database, it takes a certain amount of time to explore it, understand it and understand how it works, especially when the data are numerous and complex, which is the case in this project. What data was collected? What has been recopied exactly as the original sources? What has been noted or coded by interpretation? Is the data rigorously organized? These are some questions that had to be answered before we even thought about how we would put all this online.
Thinking about the user experience: from spreadsheet to web interface
After the expertise work of the data sources and the database comes the moment of reflection to make the data searchable via a web interface. Whether it is the hobby genealogist or university researcher, the neophyte or the regular database user, everyone must be given the opportunity to find their way around as best they can according to their practices, and it is not always easy!
Someone who seeks to establish his or her family genealogy will not have the same practices as a social science researcher who would conduct a statistical study on slavery. However, the data collected by Mr. de Reynal had not been designed, at first, to end up on a web interface. What possibilities should the search engine offer? What criteria for refining results will be relevant for one or the other? How to display the results to provide as much clear and relevant information as possible at a glance? Answering these and many other questions helps to structure the data, establish the search methods to be proposed and the way the results of the queries are displayed.
A lot of work is therefore being done in parallel with this reflection to restructure the data, code them, create concordance tables between content and codes…. In short, to prepare content for a wide variety of users. In this case, technical support by an IT professional is essential! In this matter, I salute the patience of Laurent, the software engineer, who had to (re)test and (re)program, a whole bunch of elements both for the database and for the whole web environment that surrounds it.
Thinking about the user experience: accompanying in understanding
Accessing an online searchable database is fine; but understanding its structure and content is better! To allow everyone to have the best possible user experience, it was important to think about the entire web environment around the database itself. What information does the user need to make the best use of the database? What information does it need to understand the content?
Extensive collaborative work between the project partners made it possible to produce different types of texts and to choose illustrations to present the project and the possibilities offered by the database. The exchanges also provided an opportunity to list both the technical elements to help with research, but also the historical knowledge elements to better understand the data contained in the database. I spent a lot of time writing… reading, rereading, rewriting, having someone proofread, correct, modify, add, complete, reorganize, correct again… the information notice which was supposed to both help with research, but also give elements of understanding about the data while thinking about the needs of as many types of users as possible.
Diagnosis of the existing situation to consider the future
Every effort has been made to make this database a useful and accessible tool, taking into account time and technical constraints; however, there is always room for improvement. The last step of my work in the project was therefore to make a diagnosis of what had been achieved: to establish possible improvements, but also to think about the next steps of the project to allow it to grow further. In the meantime, I invite you to discover our history with the « Esclavage en Martinique » database.
Tan Listwa 