1. Introduction

I am going to present a model of hypertextual database centred on early dictionaries, a model in large part already realized and operating on PC under the text retrieval program WordCruncher for Windows.1 The model is conceived in the first place to allow the researcher to work either at home or in an institutional setting. The dictionaries and texts easily fit on a hard disk. The ability to store images depends on their number, the degree of detail and compression techniques, all limiting factors of decreasing importance with each new technological advance.

My presentation is intended to be both technical, in order to explain the model, and philological, to show the interest of the tool. And, beyond the language dictionary, I should like to suggest the perspective of cultural knowledge databases that are increasingly the object of computer-driven projects and products. The bases linked to the dictionaries are conceived as an enhancement and an enriching of the latter. By itself the dictionary base gives access to all occurrences of a given phenomenon, while the associated bases inform and complete the dictionary data.

The model comprises several components: 1) the dictionary base; 2) the text base, containing texts chosen for having served as sources of the dictionaries: the captured editions are therefore all 16th-century ones (princeps in the case of Du Fouilloux, Vigenere and Vitruve-Martin, unique in that of the Triomphe de Henry); 3) the metalinguistic keyword base, which I shall explain subsequently; 4) the bibliographical base, containing information on the dictionaries, texts, quoted sources and commentators; and 5) image and note files. The different components are associated to each other either by hypertextual links or by the user's queries.

The early dictionary base contains the following works:

The source text base includes: I shall comment on the different elements using concrete examples. My paper includes, in its electronic form, two components: a) a technical and philological commentary, with links to 2) screen displays of elements of the different bases. Within the base context displays bolding is used to mark occurrences of the requested phenomenon; hot links are explained at some point in the commentary; the typographical distinction roman/italic is that of the quoted dictionary or text.

[Return to Table] -- [Continue]

Note 1. Johnston & Co., 1995.