Support for more language codes

ConversazioniRecommend Site Improvements

Iscriviti a LibraryThing per pubblicare un messaggio.

Support for more language codes

1gcthomas
Mar 1, 2021, 2:08 am

It would be great if LibraryThing supported the full set of ISO 639-3 or Glottolog language codes for the book language fields.

This would be particularly helpful when recording the "original language" of books that have been translated from written or oral accounts in obscure languages.

2timspalding
Mar 1, 2021, 6:49 am

Hmmm. My understanding is that the MARC language list is compatible with ISO 639-2, but I'm not sure about ISO 639-3. Are you a cataloger?

3birder4106
Mar 2, 2021, 6:29 am

I would prefer ISO 639-3.

See at https://de.wikipedia.org/wiki/ISO_639 => ISO 639-3
German:
Siehe auch: Listen der ISO-639-3-Codes:
Die Norm ISO 639-3 wurde am 5. Februar 2007 herausgegeben9 und soll aufbauend auf den ersten beiden Teilnormen eine umfassende Abdeckung aller Sprachen der Welt ermöglichen. Die Kennungen aus drei Buchstaben aus der vorhergehenden Norm ISO 639-2 werden weitergeführt und somit kann auch ISO 639-3 theoretisch über 17.576 verschiedene Kennungen verfügen (praktisch unter anderem dadurch begrenzt, dass ISO 639-5 ebenfalls Alpha-3-Codes aufnimmt, die disjunkt (elementfremd) zu denen aus ISO 639-3 sind). Aufgenommen werden alle bekannten Sprachen, worunter auch alle lebendigen, ausgestorbenen, historischen sowie auch konstruierten Sprachen fallen. Mehr als 6.900 Sprachen sind bisher in den Standard aufgenommen worden. Gedacht ist die komplette Liste vor allem für den Einsatz in der Informationstechnologie, wo eine komplette Auflistung aller Sprachen wünschenswert ist.10 Darunter sind auch Einträge wie für die schweizerdeutschen Dialekte (gsw, German SWiss),11 Kölsch (ksh) und die bairischen Dialekte (bar).

english translation:
The ISO 639-3 standard was published on February 5, 2007 9 and, building on the first two sub-standards, is intended to enable comprehensive coverage of all languages ​​in the world. The three-letter identifiers from the previous ISO 639-2 standard are continued and thus ISO 639-3 can theoretically also have 17,576 different identifiers (practically limited, among other things, by the fact that ISO 639-5 also includes alpha-3 codes that disjoint (foreign to the element) to those from ISO 639-3). All known languages ​​are recorded, including all living, extinct, historical and constructed languages. More than 6,900 languages ​​have been included in the standard so far. The complete list is intended primarily for use in information technology, where a complete listing of all languages ​​is desirable. 10 This also includes entries for the Swiss-German dialects (gsw, German SWiss), 11 Colognian (ksh) and the Bavarian dialects (bar). SO 639-3: 2007 - Codes for the representation of names of languages ​​- Part 3: Alpha-3 code for a comprehensive coverage of languages

At least the differentiation between German variants or dialects such as Bavarian, Colognian, Swiss German or Walliserdeutsch (wae) would be better that way.

4gcthomas
Modificato: Mar 7, 2021, 6:54 pm

ISO 639-2 was based on the MARC language list, which was geared toward the main languages used in world literature. ISO 639-3 adds many additional language codes for smaller languages documented by linguists, with the goal of covering all known natural languages. Some ISO 639-2 codes for groups of languages are not included in ISO 639-3.

Background on my use case: I'm using LibraryThing to catalog my personal library. One of my areas of interest is Native American literature, which is often based on oral sources in languages currently spoken by only a few hundred or thousand people. It would be nice to be able to label these by specific language, but in some cases the only available option is language family or even "Other Native American languages".

Take for example this book: https://www.librarything.com/work/490928/details/195353517

This book was published in English based on oral recordings/transcripts in Hopi, an Uto-Aztecan language with only around 6,800 native speakers. Based on the MARC guidelines (https://www.loc.gov/marc/languages/language_name.html#h), this language is coded as cai, Central American Indian (Other), but in ISO 639-3 this could be coded more specifically as hop, Hopi.

If MARC is the prevailing standard for library science it seems important to keep supporting that, but maybe there could be an option to add an additional note or supplementary language code.