EDD Online applied, corrected and supplemented

In the preceding project EDD Online a team of the English Department of the University of Innsbruck has managed to digitise the renowned English Dialect Dictionary by Joseph Wright (1898-1905) and to make it accessible in its multi-faceted linguistic details by means of an ambitious software. The six- volume dictionary is unique by its enormous size (some 4,500 pages) and its remarkable scholarly quality and complexity. It describes the dialects of the United Kingdom (incl. Ireland) as well as those of the English colonies of the late 19th century for the period from 1700 to 1904 and, moreover, provides details on the roots of English dialect vocabulary since Old English. Our interface was launched in April 2016. It has already been well received by various types of users, from English dialect societies to a large number of individual linguists worldwide. The first reactions have been overwhelming: "a wonderful tool ... brilliant ... a clearly tremendous contribution". One of the reasons for this is our ambitious philological evaluation of Wright`s dictionary, with its over 100,000 headwords, some 120,000 variants and an endless number of quotations and sources. Another reason is that we have developed a unique software, written specially for this purpose. The interface provides ten searching parameters, from full text to text in citation, phrases and variants. Moreover, it offers eight different filters, concerning place and time of a dialect word as well as its usage, but also its etymology. The dialectal distribution is illustrated on maps of either the UK or the world. Both flora and fauna are referred to in their dialect terminology and with pop-up pictures from the internet. Nevertheless EDD Online is not perfect. Given the enormous size of our software, some bugs and mistakes, noticed over the last weeks, have not come as a surprise. In the face of the international community of researchers and users worldwide we have to reckon with further references to mistakes. Our first aim in the present project is, therefore, the correction of the software. Second, for EDD Online to come closer to perfection, the 179 pages of Wright`s Supplement have to be implemented. This implies some special problems and could not be included in the preceding project, but the Supplement was part of the initially scanned text. A third aim of the present project is to apply the results of the previous project. The use of EDD Online shall be promoted both internationally, by an adequate marketing and by research cooperation, and locally in Innsbruck, by way of publications generally and the focussed support of an exemplary master thesis written by a student team member. Given the team`s expertise, the three aims can be achieved within a year.

ORD 89-VO, a follow-up project to TRP 116-G20 (2011 to 3/2016), was started on 1 May 2017, after my scrupulous preparation of our work. In addition to the two programmers, I hired two female team members, who were responsible of the correction of our XML text and of the creation of query commands. After its extension, the project ended on 31 October 2018. Our work first concerned the correction of the Supplement text. As this task had been started by myself before the official beginning of the project, it could soon be concluded (in June). The second main task was the annotation of the Supplement text (so-called tagging), which turned out to be somewhat more labour-intensive than expected because, in spite of the relative shortness of the text (179 pages), all theoretically available tags had to be considered and attributed. Partly simultaneously to this tagging we had to do correction work on the complete Dictionary text and to eliminate "bugs". However, our main aim in this phase was to implement new query tools. One of these was the creation of a kwic-concordance, another one the programming of a new filter within the menu of usage labels concerning different degrees of reliability of the EDD`s data. The work in this phase also included diverse normalisations and quantifications applied to the result lists of queries. For my theoretical preparation of this task I spent a day at the University of Salzburg with my colleague Prof. Hans Goebl, who is a most qualified specialist in matters of geostatistical quantification of dialect data (so-called dialectometry). Moreover, we hosted Dr. Heinrich Ramisch, a colleague of the University of Bamberg, for a guest lecture and talks in Innsbruck, thus increasing our competence in the domains of quantification and dialect cartography. The bugs we were after and managed to eliminate mainly concerned the precise identification of dialect areas and the diverse labels created by us. One such bug was found in the abbreviation Cor*, which turned out ambiguously to refer to both `Cornwall` and Shakespeare`s play `Coriolanus`. We were faced with many such ambiguities. Moreover, the more than 1000 sources of the EDD as well as the logics of the compatibility of filters caused problems which had to be solved. Beyond our interface we also cared for the distribution and marketing of EDD Online. All websites concerned have been revised. In 2017, I published a new online version of the EDD Manual and registered the project and my/our publications concerning the EDD on the internet platform researchgate. There have been four external lecturing visits of members of the project. In terms of publications, we wrote seven project-related articles and reviews. Having been concerned with the EDD since 2003, I can now, in hindsight, say with some satisfaction that EDD Online in its present form has by far surpassed all my earlier expectations. The more I would like to express my sincere thanks to all team members and the functionaries of the University of Innsbruck as well as the Austrian Science Fund in Vienna for their interest and solidarity with which they have supported the project and my own work over so many years.