The main goals of the project are:
the development of advanced transliteration software for the conversion of bibliographic records into and out of the Greek character set;
to promote the wider use of consistent and reversible transliteration standards.
Machine-readable records for Greek language materials exist mostly in transliterated form. Past practices in transliteration have been diverse and the use of reversible schemes limited, thus impeding the conversion and exchange of records for use in the Greek alphabet. The project will develop software to aid the conversion of bibliographic records between Greek & Roman alphabet languages. The structure of the bibliographic record, the MARC format, and the relationship of key data elements to authority files will be used to enhance the transliteration software. In the process, research will be carried out on name authority issues and transliteration standards and practices. The software will be designed as general purpose modules for use in a stand-alone environment but will be tested operationally by being integrated into a library system in Greece.
Impact and expected results:
The project gives improved and much wider access to Greek materials for both native and non-native Greek speakers and greatly encourages the exchange of records for current and retrospective cataloguing
It has also raised awareness of standards issues in relation to reversible transliteration schemes. In addition, it has made a significant contribution to the understanding of transliteration problems and has advanced solutions which can be applied to other alphabets.
Deliverables are available in electronic and paper versions. The electronic version includes core and applications software for integration into the libraries environment. With it are two guides: one for users and the other for developers wishing to integrate and utilise the software in libraries applications. Software design, specification and testing documents are project deliverables but are confidential to it.
In total, publications cover:
HELEN Users Guide;
HELEN Developers Guide;
Transliteration principles and practice;
Transliteration and MARC;
PC character sets;
The project was structured into six related modules:
Transliteration and MARC;
As well as the development of modules and their integration as advanced transliteration software, HELEN went on to use its results to promote the wider use of transliteration standards. Both endeavours were served by the successful installation and testing of the software in an operational environment at the University of Crete.
The basis for development was a core transliteration module to handle simple character by character transliteration, logically independent of any character-set representation scheme. This was enhanced in stages: to handle main, common transliteration schemes in different languages; to incorporate name conversion routines, based on name authority files; to develop more intelligent features such as automatic recognition of transliteration schemes, routines for treating ambiguous characters. Problems of access and display of Greek script records in Roman alphabets were investigated and solutions tested. The software was evaluated and tested by partners in Greece after each of the key development phases as input to the final version.
The main technical issues to be explored were:
- Reconciliation of past transliteration practices with current requirements for easy reversible transliteration;
- Use and integration of name authority files as a tool in transliteration;
- Development of transliteration tools for potential multi-lingual sources;
- Conformity with standards for character sets, bibliographic records in Greek environment.
Further information available from contact below and from http://alcyone.cc.uch.gr/~kosmas/Helen/index.html or from ftp://ftp.comp.brad.ac.uk/pub/library/helen.