Language Return to XOBIS main page

Structure of Language Principal Element

2.Watch Your Language!(Linguistic Records)

Language

Specific spoken, written, or signed communication systems

<Language>
    <Entry
        <Name>Esperanto</Name>
    </Entry>
</Language>

The Language element refers to specific named languages, which are defined as verbal or nonverbal systems used by beings to express ideas and feelings. For working purposes, it encompasses spoken, written, and signed languages, as well as specific named dialects, subsets, and usages. Also in scope are artificial and fictional languages.Languages are instantiations of Concept, usually language families. Individual words and phrases of Language are instantiated using the Principal Element String. Currently, the optional 'usage' attribute is defined for Language, with the single value subdivision, indicating that it may be used as the value for Subdivision.See Entry Substitutes under the Generic Elements section above for accommodation of abbreviations and codes.See 'language' in the Attributes section for intra-record linguistic designations.

Language excludes computer programming languages, markup languages, scripts, shorthand systems, alphabets, fonts, etc., which are most likely authorities or instances of Work. These can be linked as necessary; see the Work section for further explanation.Also excluded are linguistic groups that employ a proper adjective, e.g. Indic languages, instead of proper noun, and are thus covered by Concept. This is consistent with XOBIS' reliance on the same technique as a guideline for determining cusps. See the Indexing section for combining entries from different Principal Elements.

The Entry for Language has four attributes:'scheme' with value from Code of a Work, 'class', with working values: individual, other, and referential, 'language', and 'transliteration'.The value individual permits the current set of single languages allowed for assigning to works, while other covers dialects, etc. and referential covers informational entries.

Relationships are used to categorize languages and indicate other associations. Linguistic Records permit conceptual relationships in the form of categorical linking to record that a Work is in a particular language ("isness") and topical linking to treat a language as a subject ("aboutness").This keeps a language and its literature discrete. There are many possibilities, e.g. geographic relationships as a basis for a linguistic map interface; vital relationships for identifying available translators, scholars of rare languages, and writers by language; organizational relationships for official languages; chronological relationships for when languages were extant; linguistic relationships to other languages (e.g. Old English being continued by Middle English); etc.

A Relationship may have an equivocation via its Modifier, e.g. "chiefly" British [usage].This could also be used to indicate the fluency of a person's reading, writing, and/or speaking ability, although the 'degree' attribute is a potential alternative. Suggested categorical relationships and selected others are shown in these examples using italics:

Language Category:(Singular of Concept)
Afrikaans Germanic Language
American Sign Language Sign Language
Related: English Language
Basque Language Isolate ?
Elvish Language, Fictional ?
English Germanic Language
Earlier: English, Middle Extinct Language ?
Narrower or Related:
American Sign Language Sign Language
Basic English Language Subset ?
Black English Language Usage ?
English (American Usage) ? Language Usage ?
English (British Usage) ? Language Usage ?
Esperanto Language, Artificial
Flash (Prisoners' Dialect) Dialect / Slang
Related: Prisoners Concept
Related: Australia Place
Related: English Language
Sea Islands Creole Dialect Creole Dialect
Yoruba Kwa Language

The World Wide Web indicates the need for internationalization in its name. Language-specific versions of software have resolved the problem only partially.Much data remains in one language or lumped in the same record without an indication of language. Worse yet, cataloging too frequently mixes languages in the same entry, either deliberately or from lack of information.XML offers an excellent foundation for addressing such problems due to its use of Unicode (54) for flexible character set support without special provisions. As an initial step in an effort to address the data issue in XOBIS, 'language' and 'transliteration' attributes are available for Entry, Varia, and a number of other elements, cf. the Attributes section. We elected to resect an initial language clustering attempt in favor of further study and, hopefully, much needed input from the international community.The issues are nontrivial due the need to correlate Entry/Varia, their Qualifiers, and the references in Relationships to specific IDs.We envision user-selectable language preferences that would rely on a default mechanism when entries or references are not available in the selected language for a particular value. As a result, importing records using a different primary language should be easier to incorporate into a local file using a different primary one.We expect a beta version of XOBIS to benefit from further investigation in support of more sophisticated multi-lingual catalogs.