On 02/10/2013 12.21, Heidrun Wiesenmüller wrote:
I recently taught at RDA at the National Library of Israel. They do not have a single preferred script, nor a single language of cataloging. In fact they have four: Hebrew, Arabic, roman, and Cyrillic. Depending on the script of the resource they are cataloging, they will use an authorized access point in that script and the language of cataloging will depend on the language of the resource. They have a unique authority record structure which uses a single record with multiple 1XXs for the authorized form in different scripts.
I’m not a specialist for original script cataloging, but I can try and explain (at least roughly) what is being done in Germany. I’ll take my own union catalog, the Southwest German Library Network (SWB), as an example.
In the title records, the most important fields are duplicated: One version of the field is then used for the transliterated text and the other for the same text in the original script. A code for the script is recorded in a subfield. Note that the format for the title records is not MARC.
If you want to see what this looks like in the catalog, try this:
Interesting. You can see the National Library of Israel practice in VIAF, e.g. for Leo Tolstoy http://www.viaf.org/viaf/96987389/#Tolstoy,_Leo,_graf,_1828-1910 and Tolstoy has several valid forms of heading. In fact, you can compare the National Library of Israel cyrillic form vs. the Russian form. Of course, both are in cyrillic, but the Russian form does not have “граф” (graf, or Count).
There are other options however. At the Food and Agriculture Organization of the United Nations, they have authorized forms but they vary depending on the language of the item. They publish in 6 languages and when something is in English, they use the English form, in Spanish, the Spanish form and so on. Here is an example: bit.ly/1eZeOrO with a corporate body form of “FAO, Rome (Italy). Rural Infrastructure and Agro-Industries Div.” If you look toward the bottom of the record, you’ll see “Rel. lang. versions”. Click on French and you’ll see “FAO, Rome (Italy). Div. des Infrastructures Rurales et des Agro-industries” or Spanish, “FAO, Rome (Italy). Direccion de Infraestructura Rural y Agroindustrias”. There are other practices as well.
The reality for our users is changing from single authorized forms to multiple authorized forms, not only as we see in the Israeli library and FAO, or even in VIAF, but probably even more available to the public will be something as we see in dbpedia http://dbpedia.org/page/Leo_Tolstoy. That is, if the public ends up using authorized forms at all or if we expect computers to do everything automatically.
If they are going to use these forms, based on URIs or whatever, the real problem will not be technical since that is relatively simple now (and that is an amazing fact!), but the real problem will be practical: how can we take this huge amount of content and make something coherent and comprehensible for people who will refuse to spend 20 hours in a searching workshop?
That’s a challenge.