Baybars Hackathon – Call for applications – Beirut, Sept. 26-29 2023

French Institute of the Near-East (Ifpo)
Beirut
September 26-29 2023

[بالعربية هنا]
[En français ici]

As part of the research program “Popular Literatures of the Levant » (Littératures populaires du Levant – LiPoL) and in partnership with the National Library of France (BnF), the startup company Calfa and the media library of the Mediterranean House of Human Sciences (MMSH), the French Institute of the Near-East (Ifpo) is organizing from September 2023 in Beirut a hackathon to transcribe a corpus of manuscripts from the Sīrat Baybars in its Damascene recension. This corpus consists of 542 storytellers’ notebooks corresponding to around 60,000 pages of text in Middle Arabic. A third of the corpus has already been transcribed for publication by the Presses de l’Ifpo.

This “Baybars Hackathon” aims to obtain an HTR (Handwritten Text recognition) model for the automatic transcription of the entire corpus, thanks to the initial manual transcription of a sample of the manuscripts.

The project will involve a team of 12 people selected through an application process from Master’s and PhD students enrolled in a Lebanese university and specialized in Arabic language and literature, History or codicology. Students from French universities who are already members of the LiPoL program will also attend the hackathon.

A preliminary four-day training will take place at Ifpo Beirut from September 26 to September 29, 2023. This training will address a range of issues relating to Arabic-language manuscripts in general, and to manuscripts of popular literature in particular, and to the digital humanities, around four themes:
– HTR training provided by Calfa, a company specialized in OCR and HTR for non-Latin languages, which will support us throughout the project;
– an introduction to Arabic manuscripts through codicology and palaeography;
– a training in Middle Arabic, a register of Arabic language situated between standard and dialect, widely used by storytellers of popular and middle Arabic literature, which the sīra is part of;
– an introduction to popular and middle Levantine literature, to the handwritten sources of this literature, to the sīra shaʿbiyya genre and Sīrat Baybars.

At the end of the training days, participants will leave with a set of pages to transcribe on their own during the following weeks in order to train the Artificial Intelligence which will generate the HTR model.

The “Baybars Hackathon” is part of a range of numerous scientific initiatives which aim to improve the place of Arabic language in the Digital Humanities field and in the scientific research. It follows the “Alexander Hackathon”, organized between November 2022 and May 2023 around manuscripts of the Alexander Romance in Arabic. The Alexander Hackathon produced textual data which will serve the critical edition of the romance. The event involved students from Sorbonne University, Strasbourg University, Lorraine University and Aix-Marseille University (see: https://lipol.hypotheses.org/790).

Partnership

The “Baybars Hackathon” is organized as part of the research program “Popular Literatures of the Levant » (Littératures populaires du Levant – LiPoL), which is funded by the French National Research Agency (ANR), in partnership with the National Library of France (BnF) and the Mediterranean House of Human Sciences (MMSH).

Technical and scientific supervision

Taos BABOUR, Research Engineer, head of the Digital Humanities department at the Ifpo, Beirut; email: t.babour[at]ifporient.org
Claire CIALONE-GREGOIRE, Digital Humanities Engineer for LiPoL program; email: claire.cialone-gregoire[at]cnrs.fr
Iyas HASSAN, University Professor in Arabic Literature at Sorbonne University and member of the Ifpo, head of LiPoL program; email: iyas.hassan[at]cnrs.fr

Application and validation of the participation

The Baybars Hackathon is open to Master’s and PhD students enrolled in Lebanese universities in Arabic language and literature, in History, codicology, or any other subject related to Arabic studies and Arabic handwritten sources.

The workshop is open to beginners and no previous experience with manuscripts, HTR, popular literature or Middle Arabic is required. However, perfect knowledge of written standard Arabic and its grammatical rules is mandatory to participate in the hackathon and benefit from the training.
The applicant shall provide a recommendation letter from his/her course supervisor, explaining how the participation to the hackathon could benefit his/her university education.
At the end of the hackathon in 2024, participants will receive a certificate of attendance from the Ifpo. With the prior agreement of each university’s department, participation in the hackathon could be validated as part of the student’s course.

A commitment from the participants will be required for the entire project, from September 2023 to February 2024 (see provisional schedule below), in order to regularly check the quality of the transcription produced by the HTR.

Shared room accommodation at the Ifpo may be available for participants from universities outside Beirut (limited places).

Provisional schedule

September 8, 2023: Application deadline
September 26-29, 2023: Initial training at Ifpo Beirut
September 29 – October 31, 2023: Transcription work (remotely)
October 31, 2023: Submission of all samples
November 7, 2023: Verified transcripts will be sent to Calfa for processing
February 1, 2024: Reception of the text transcribed by the HTR model developed by Calfa

In order to apply to the Baybars Hackathon

Please send your resume with the recommendation of your course supervisor to :
Taos BABOUR – t.babour[at]ifporient.org
Claire CIALONE-GREGOIRE – claire.cialone-gregoire[at]cnrs.fr
Iyas HASSAN – iyas.hassan[at]cnrs.fr

First published on: https://www.ifporient.org/baybars-hackathon-call

Featured image: first page of the notebook number 12 from Abu Ahmad’s manuscript of sīrat Baybars (baybars-nt-ms1-l012-003), Damascus, 1949