Periodic Reporting for period 1 - DigiFlor (Digital Edition of the Roman de Florimont)
Periodo di rendicontazione: 2017-05-01 al 2019-04-30
The project aimed to produce a digital scholarly edition of an overlooked Medieval text: the Roman de Florimont (1188). Almost unique among the contemporary romans, it offers a Mediterranean setting: its hero moves between Greece, Southern Italy, Albania, etc. and its author declares to have translated it from a Greek original. Furthermore, this thematic interest intersects with a remarkable linguistic relevance as the tradition includes witnesses related to two understudied linguistic domains: Franco-Provençal (the third French language, together with the Langue d'Oil and the Langue d'Oc) and Franco-Italian (an artificial literary language). Franco-Provençal counts much documentary evidence but very few literary witnesses; therefore, the recovery of a poetic text within this domain constitutes a very important contribution to the field. The focus on the Franco-Italian manuscripts can be understood in the context of the growing interest for the Medieval European and extra-European diffusion of French as a literary language. Despite this interest, Florimont has been almost entirely neglected by the scholarly community: the only existing edition (Hilka’s in 1932) is seriously flawed.
The main objective of the project was the creation of a “complex” digital edition aiming to offer “something more” than a printed edition. First of all, to the display level; the editorial model was a diplomatic edition (i.e. an edition faithful to the manuscript) offering all those details that are usually waived, especially due to the physical limits and to the costs imposed by the paper book format, above all in presence of several manuscripts; digital technology allows to overcome this obstacle, with the further advantage of considerably shortening publication times. But also at the level of “analysis potential”: the second advantage of a digital edition is in fact the possibility to manipulate the text, carry out research, extract statistical data and add hypertextual links.
The project therefore represented a moment of critical reflection around the digital editorial practices concerning above all the manuscript texts; and it ended with a series of alternative proposals having the ambition to solicit discussion. Secondly, the project has made available to the academic community some prototype tools that want to represent a stimulus to discuss some current practices:
The main objectives of the project were the production of the Roman de Florimont edition (four manuscripts) and the study and test of digital philological tools applied to Medieval vernacular texts.
The main objective of the project was the creation of a “complex” digital edition aiming to offer “something more” than a printed edition. First of all, to the display level; the editorial model was a diplomatic edition (i.e. an edition faithful to the manuscript) offering all those details that are usually waived, especially due to the physical limits and to the costs imposed by the paper book format, above all in presence of several manuscripts; digital technology allows to overcome this obstacle, with the further advantage of considerably shortening publication times. But also at the level of “analysis potential”: the second advantage of a digital edition is in fact the possibility to manipulate the text, carry out research, extract statistical data and add hypertextual links.
The project therefore represented a moment of critical reflection around the digital editorial practices concerning above all the manuscript texts; and it ended with a series of alternative proposals having the ambition to solicit discussion. Secondly, the project has made available to the academic community some prototype tools that want to represent a stimulus to discuss some current practices:
The main objectives of the project were the production of the Roman de Florimont edition (four manuscripts) and the study and test of digital philological tools applied to Medieval vernacular texts.
The resuts:
a. the creation of a very complex digital diplomatic edition, taking into account all the peculiarities of Medieval writing; all these encoded phenomena can be the subject of a search within the text, allowing the extraction of statistical data;
b. the creation of an editor XML-prototype allowing a complete automation of the XML-TEI files production; this editor allows the philologist to work with a simple .txt file and, in this way, to concentrate on the text and on the work of transcription and analysis;
c, the creation of the interface for the visualization of the editions and of the site structure.
Thanks to the tools created during the project, the textual corpus has been further enriched: from four to six editions for a total of 81.000 verses.
Dissemination:
a. project results are visible through the site DigiFlorimont. Archive numérique du “Roman de Florimont” d’Aimon de Varennes http://digiflorimont.huma-num.fr/ ;
b. an academic blog was activated to offer theoretical reflections concerning the application of digital technology to Medieval textuality: Florimont numérique. Textualité médiévale et textualité numérique ISSN 2646-0742;
c. all the files and tools produced will be stored in the GitHub repository https://github.com/digiflor;
d. the project obtained the affiliation to the CAHIER consortium https://cahier.hypotheses.org/;
e. a key moment for the dissemination was the organization of the international conference Autour de Florimont. Textualité médiévale et textualité numérique (Grenoble, 13-14 December 2018), that brought together 15 speakers from Canada, United States, England, Belgium, Germany, Italy, and France.
a. the creation of a very complex digital diplomatic edition, taking into account all the peculiarities of Medieval writing; all these encoded phenomena can be the subject of a search within the text, allowing the extraction of statistical data;
b. the creation of an editor XML-prototype allowing a complete automation of the XML-TEI files production; this editor allows the philologist to work with a simple .txt file and, in this way, to concentrate on the text and on the work of transcription and analysis;
c, the creation of the interface for the visualization of the editions and of the site structure.
Thanks to the tools created during the project, the textual corpus has been further enriched: from four to six editions for a total of 81.000 verses.
Dissemination:
a. project results are visible through the site DigiFlorimont. Archive numérique du “Roman de Florimont” d’Aimon de Varennes http://digiflorimont.huma-num.fr/ ;
b. an academic blog was activated to offer theoretical reflections concerning the application of digital technology to Medieval textuality: Florimont numérique. Textualité médiévale et textualité numérique ISSN 2646-0742;
c. all the files and tools produced will be stored in the GitHub repository https://github.com/digiflor;
d. the project obtained the affiliation to the CAHIER consortium https://cahier.hypotheses.org/;
e. a key moment for the dissemination was the organization of the international conference Autour de Florimont. Textualité médiévale et textualité numérique (Grenoble, 13-14 December 2018), that brought together 15 speakers from Canada, United States, England, Belgium, Germany, Italy, and France.
From a philological point of view, the project has brought to the attention a very little known text of Medieval French literature, that represents an important piece in the context of the literature culturally linked to the novels of antiquity. Its manuscript tradition is characterized by remarkable linguistic variation, textual variants and a wide geographical distribution and therefore represents an exceptional example for the theoretical reflection on the concepts of: variant, critical edition and documentary edition. The complexity of the problems related to this tradition means that, beyond the results obtained in this first phase, the project is open to further developments: the next step will be the elaboration of a critical edition model.
From a technical point of view, the implemented encoding system presents a series of advantages:
a. it is easily manipulated in an IT perspective because the text is completely formalized and therefore machine-readable;
b. it will allow to experiment the most complex encoding system for a critical edition, which is regarded as the best methodology but is not applied for lack of adequate tools to produce the code;
c. it will also facilitate the enrichment of the edition with a linguistic encoding.
The main problem in creating such a complex encoding is the lack of ergonomic tools allowing to realize this code within a reasonable time, standardizing the production and then multiplying the examples: the risk is indeed to produce complex but isolated prototypes. The prototype-editor developped doesn’t have the ambition to present itself as the tool par excellence, but it represents at least a concrete example of an alternative to current practice. The most widespread tools don’t allow the realization of a so complex encoding in a reasonable time because the code is completely typed by the encoder. In this regard, one can often observe a discrepancy between the official declarations and the practice in the realization of the projects: often, especially with large textual corpora, each team works with some personal solution to automate at least part of the encoding process. The digital editions production is difficult, the problem is known but there is no official discussion about.
All the technical solutions adopted represent innovations that it would be desirable to engineer and test in different contexts, to be able to make solid tools available to the community. In particular, the operating principle of the editor, currently from .txt to .xml, could be the basis of an implementation allowing a simplified conversion from .xml (the encoding language) to .html (the web visualization language). This conversion represents a very problematic moment in the process of creating a digital edition, and it is currently based on a very complex and now obsolete language, .xslt. These tools would interest any humanist engaged in the production of a digital edition, especially in the case of a very complex encoding, and whatever the nature of the text: the field of the tools available to the digital humanists is still open, and the demand for efficient tools is very strong.
From a technical point of view, the implemented encoding system presents a series of advantages:
a. it is easily manipulated in an IT perspective because the text is completely formalized and therefore machine-readable;
b. it will allow to experiment the most complex encoding system for a critical edition, which is regarded as the best methodology but is not applied for lack of adequate tools to produce the code;
c. it will also facilitate the enrichment of the edition with a linguistic encoding.
The main problem in creating such a complex encoding is the lack of ergonomic tools allowing to realize this code within a reasonable time, standardizing the production and then multiplying the examples: the risk is indeed to produce complex but isolated prototypes. The prototype-editor developped doesn’t have the ambition to present itself as the tool par excellence, but it represents at least a concrete example of an alternative to current practice. The most widespread tools don’t allow the realization of a so complex encoding in a reasonable time because the code is completely typed by the encoder. In this regard, one can often observe a discrepancy between the official declarations and the practice in the realization of the projects: often, especially with large textual corpora, each team works with some personal solution to automate at least part of the encoding process. The digital editions production is difficult, the problem is known but there is no official discussion about.
All the technical solutions adopted represent innovations that it would be desirable to engineer and test in different contexts, to be able to make solid tools available to the community. In particular, the operating principle of the editor, currently from .txt to .xml, could be the basis of an implementation allowing a simplified conversion from .xml (the encoding language) to .html (the web visualization language). This conversion represents a very problematic moment in the process of creating a digital edition, and it is currently based on a very complex and now obsolete language, .xslt. These tools would interest any humanist engaged in the production of a digital edition, especially in the case of a very complex encoding, and whatever the nature of the text: the field of the tools available to the digital humanists is still open, and the demand for efficient tools is very strong.