ИСТИНА

In the report, we discuss various constraints imposed on lexicographic databases being developed in long-term projects aimed at description of an underresourced language and publishing the dictionary of the language both in printed form and as an online resource. We present the dictionary database for the Beserman dialect of Udmurt, collected during an ongoing fieldwork project started in 2003. The paper describes the challenges that arise in projects of such kind and the proposed solutions. For the dictionary database, we used the TshwaneLex software which provides a reasonable balance between flexibility and customizability (needed for dealing with the chalenges), on the one hand, and orderedness of the data (needed for seamless interaction with the online interface), on the other. A preliminary print version of the dictionary based on the presented database was published in 2013 (Kuznetsova et al. 2013). An online searchable web interface based on the database and a corpus of oral Beserman texts, is currently under development and is expected to become available for public access by May 2015.