I propose the data tags containing names (name tag) street, square, historical site names, be standardized to Latin alphabet (latinica) instead of Cyrillic alphabet.
I realize that this is contrary to the official country guidelines (and contrary to the charter of certain language purist groups), however there are several advantages to using latinica at this time, not least of which is that it enables search functionality on devices that don't have Cyrillic input methods set-up, such as most smart phones / mobile devices. Also, most indexers today only look at basic name tags, and using it helps with text to speech engines, and it's easier to remember, visually match and spell for visitors that don't use Cyrillic alphabets in their own language. I don't believe any of the user's will have trouble with names in latinica.
I propose the use of name:rs (extended tag with country's two-letter code), for Cyrillic name in Serbian language and name:en for English name where applicable (shouldn't contain diacritics). Places with names such as "British Council" or "McDonald's", the spelling should be left intact, contrary to Serbian language rules (not Britiš kaunsil, Mekdonalds), for the same reasons that were mentioned above.
For alternative naming (see: Slavija vs. Trg Dimitrija Tucovića), I propose the use of loc_name tag where official name is annotated in the name tag, and local colloquial in loc_name.
I suggest merging this page with Serbia so we could avoid any unnecessary duplication.