Conclusion and Perspectives

Having come to the end of this book, let us attempt to sum up, in a few paragraphs, the main points and issues involved in our work.

Faced with the proliferation of digital audiovisual data, one of the crucially important questions which arise is that of knowing how to turn these data into (intellectual) resources sui generis. In other words, the fact that an audiovisual document is available in digital form does not necessarily mean it becomes a (cultural) good for a given audience!

Indeed, it is one thing to set up a digital video library (or media library) containing a filmic or sound collection which, e.g. documents the activities of a scientific or teaching institution; it is quite another to transform that collection into resources for research, teaching, valorizing the institution’s image, etc. These two activities are completely dissimilar, and as long as we consider that mere digitization and uploading of such documents is a satisfactory solution to the constitution, communication and exploitation of bodies of knowledge heritage, we shall be treading a false path.

In this book, we have attempted to give both a theoretical and practical/technical treatment of the question of transforming a “simple” piece of digital data into a resource for a certain audience or user. We interpret this transformation as a process of appropriation of a piece of digital data by the audience or individual user in question. The appropriation may relate either to the materiality of an existing piece of digital information (the user modifies this, by way of montage for example), or to its content (lato sensu). This second form of appropriation is called analysis* in the broadest sense of the term: the user – the audience – describes, interprets, comments on, amends, adapts, etc., the content of the digital piece of data or a part thereof (a sequence, a segment).

Throughout this book, therefore, we have concentrated on the issue of analyzing an audiovisual text or corpus. We briefly outlined the theoretical framework which guides us, which determines our approach to the analysis of the (audiovisual) text. We also attempted to identify different types of analysis such as, e.g.: thematic analysis*, i.e. analysis of the content in the strictest sense of the term (the meaning, the message, etc.) of an audiovisual text; audiovisual analysis*, i.e. analysis of the forms of visual and/or acoustic expression in an audiovisual text; or indeed pragmatic analysis*, i.e. analysis of the adaptation of an audiovisual text to a particular use context.

Then, we attempted to define and develop a set of so-called metalinguistic resources which we need to specify and develop the models of description* needed for this-or-that type of analysis. Finally, focusing on the particular audiovisual corpora making up an audiovisual archive, we elaborated specific libraries of models of description, working on the hypothesis that every library of models describes, more or less well, more or less incompletely (or partially, i.e. according to an analyst’s own point of view and interest) the universe of discourse* of a given archive.

The conceptual organization underlying the approach put forward in this book is in fact fairly complex, and relies on a whole series of specific metalinguistic resources: meta-lexicons of conceptual terms denoting the analytical objects in the universe of discourse of an archive; a meta-lexicon denoting the analytical activities for describing a particular object or configuration of objects; a library of schemas of indexation; a library of generic building blocks for constructing models of description; libraries of models of description; thesauruses; etc.

However, the tradition of research into the text, its analysis, interpretation and “rewriting”, demonstrates that this is a complex issue which cannot be reduced to a single question, devoid of interest and very well known, which is that of indexation by keywords or delegated to the hypothetical revolutionary advanced procedure of automatic indexation which should regulate something we believe to be far beyond its range of capability. For example, human interest (curiosity, etc.) which precedes every analysis, and the hermeneutic dimension is specific to every activity of textual analysis.

This said, we are also perfectly well aware of the limitations of our approach, at least as it is presented in this book. The two meta-lexicons which constitute the “beating heart” of the ASW metalinguistic system are not yet stabilized, and it remains problematic to use them to define and develop models of description enabling us to analyze different universes of discourse than that which our research has focused on. The library of sequences of analysis of discourse production is as yet relatively poor – as is that of the sequences reserved for describing the verbal or audiovisual expression. In addition, the models of description available to us for carrying out paratextual analyses (in the vein of the Dublin Core standard) are very fixed and difficult to adapt – as indeed are those currently at our disposal for carrying out a pragmatic analysis and adaptation/translations of audiovisual data.

Other limitations to our approach relate to an as-yet fairly incomplete consideration of the “external” approaches and techniques likely to provide new solutions. For instance, one thinks of a much more advanced integration of the different standards into the ASW approach, the systematic coordination of our resources with other ontologies and thesauruses, and the use of speech to text and text mining techniques to contribute both to the enrichment of our metalinguistic resources and to a sort of pre-description/pre-indexation of audiovisual texts using the ASW models of description.

Finally, our interest in the conception and creation of a system for analyzing audiovisual corpora has always been driven by the desire to make such a system available for use by any actor, individual or collective, particularly in the “academic” world. The driving idea – so to speak – which guided us throughout these 12 years of research and development devoted to audiovisual archives, was of constituting a digital Campus of audiovisual archives for research and teaching – a sort of “academic YouTube”, a cooperative made available to anyone interested in the process of the constitution, analysis, diffusion and sharing of cultural or scientific heritage in the form of personal or collective audiovisual archives.

Our desire is, of course, to propel our research in the directions cited, and ensure the interested parties profit from it. However, as we know, this type of research cannot be done simply. It requires not only significant human, technical and financial resources but also an institutional framework which is propitious for scientific research, a certain “tranquility of spirit” – conditions which we fear have become utopian in today’s world…