User:Robert Ullmann/SC recovery/prop

(written as a precise specification)

Policy for Bosnian, Croatian, Montenegrin and Serbian


 * Bosnian uses ISO code bs.
 * Croatian uses ISO code hr.
 * Montenegrin uses wikt extension code zls-mon, will use ISO code when assigned.
 * Serbian uses ISO code sr.
 * (The language name "Serbo-Croatian" is not used. The deleted ISO code sh is not used.)
 * References to Serbo-Croatian in etymologies use ISO code hbs.
 * For Serbian, the primary entry is Cyrillic.
 * For Croatian, the primary entry is Latin.
 * Other language sections may exist, and may not be deleted (unless the word does not exist in that language, in which case the section must go to RfV or RfD process).
 * Each language section should, eventually, have examples, citations, usage etc in that language.
 * In translations sections, the 3 (or 4) languages have separate lines, not grouped.
 * Croatian and Bosnian show the translation in Latin script
 * Serbian shows the translation in Cyrillic script, with Latin as the transliteration, but linked as a proper term.
 * examples
 * Bosnian: juli, jul
 * Croatian: srpanj
 * Serbian: јули, јул
 * The example in WT:ELE policy to be changed to remove the multiple line format for Serbian.
 * A bot operation to be run to duplicate existing "Serbo-Croatian" L2 headers to Serbian for Cyrillic titles and Croatian for Latin titles, with all entries that appear to contain notes re one or more of the languages tagged for attention.
 * The bot operation will change the codes used to the appropriate codes and change the templates correctly.
 * The bot will restore missing language sections from entry history.
 * A separate bot operation will reformat the Serbian translations lines where possible, tagging the remainder in some way.

In summary, this provides full standards compliance (if SC/"sh" not used), uses the standards to follow NPOV here, avoiding taking a political position. It resolves all technical issues, supporting standard protocol with browsers and other applications. It reduces duplication of content, except where necessary between script forms, and where desired to explain language differences.