In setting standards for many products, the International Organization for Standardization (ISO 1951: 2007) deals with monolingual and multilingual, general and specialized dictionaries. It aims to establish a model for dictionary making, LEXml. It specifies a formal generic structure independent of publishing media and proposes a means to present entries in printed and electronic dictionaries in a digitalized way, including dictionary compiling, editing, publishing, and distributing. The objective of this standard is to facilitate the production, merging, comparison, extraction, exchange, dissemination, and retrieval of lexicographical data in dictionaries.

This processing model gives consideration not only to the methodical formal structure that is required for the automation of dictionary making, but also to its convenience for use; it not only introduces new ideas and new methods into lexicographic data processing but also takes into account the conventional stylistic layout and methods of the dictionary, which can be applied to both electronic/Internet dictionaries and printed dictionaries. The International Standard has the following main features:

1. Uniform framework and data items for microstructure: no matter what type of dictionary, work can be done on the basis of the same tree structure; the differences between large and small dictionaries, multilingual and monolingual dictionaries, or general and specialized dictionaries lie in the types and numbers of data items.

2. Separation of the compiling format from the display format: the format for the lexical data input is different from that for display; what compilers have to do is to input information according to the tree structure. All the punctuation marks and structure marks can be automatically generated during the phase of data display, and will be displayed on the preview interface.

3. Clear marking of the relations among microstructure information items: in the input process, imported data will be automatically tagged, and checked when they are accessed or put in use. In this way, all sorts of data can be retrieved according to the compiler's needs, which contribute to a 'smart search'.

4. Standardization of formats for all the information or data: the database is built in LEXml format, and the lexicographic data may be exported according to specific needs and be connected to the interface of special typesetting systems, or be applied to other language database, language-processing, or machine-translation systems.

5. Flexibility in data retrieval and display: the LEXml format is a universal model, and the data structure of subsets can meet specific needs as long as they are constructed according to the international standards of the XML format.

6. Good compatibility with current XML tools: if special modes and tools are used in language processing that will lead to incompatibility with other software tools. XML, along with its specifications, has become an industrial standard, and naturally the XML-based LEXml can be put into use as XML models are, and can even be edited with XML editors and XSL style sheets.

责任编辑：admin